SAP Community - Document Information Extraction

New in SAP Mobile Start V2.4: Integrate and Launch SAP Document AI on Mobile

2025-08-11T08:00:00.115000+02:00

With the SAP Mobile Start 2.4 release, users can leverage SAP Document AI—integrated seamlessly within the Mobile Start app—to transform document handling into a secure, intelligent, and automated experience. They can scan or upload documents directly from SAP Mobile Start, which are then processed by the SAP Document AI service. This streamlines processing, cuts down manual work, and boosts data accuracy throughout your operations. In this blog, I'll walk you through the integration steps, prerequisites, and configuration details needed to bring SAP Document AI capabilities to your mobile users.

Preconditions

Before you begin, ensure the following prerequisites are fulfilled.

Subscribe to SAP Document AI

Ensure that your SAP BTP subaccount is subscribed to the Document Information Extraction service (document-information-extraction-application-ias with embedded_edition instance). Once complete, access the SAP Document AI workspace from your BTP subscriptions to configure document types and schemas relevant to your business scenarios. For more information, refer to the official user guide.

Enable Identity Authentication for SAP Build Work Zone

In SAP Build Work Zone, navigate to Site Manager > Settings, and ensure Identity Authentication is enabled.

Screenshot: Enabling Identity Authentication in SAP Build Work Zone Settings

Note: Valid only for SAP Build Work Zone subscription made before 20 March 2025.

Configure Trust Relationship for API Access

To allow SAP Build Work Zone to consume SAP Document AI APIs, you need to add a dependency in the SAP Cloud Identity Services (IAS) administration console:

Go to the Applications tab
Select the SAP Build Work Zone application
Under Dependencies, add a new entry with:
Name: documentai (or any unique name)
Application: Select the “SAP Document AI” entry in your IAS tenant

Create Destination for SAP Mobile Start

To authorize SAP Mobile Start App to access SAP Document AI, create a destination in your BTP subaccount with the following settings:

Name: sapdocumentaimobilestartintegration
Type: HTTP
Authentication: NoAuthentication
Proxy Type: Internet
URL: (Use https://<your-tenant-url-endpoint>/document-ai/v1, located in Document AI instance > service key > endpoints > backend > URL)

Add the following additional properties:

HTML5.DynamicDestination = true
HTML5.IASDependencyName = documentai (dependency name created in SAP Cloud Identity Services)

Once completed, both SAP Build Work Zone and SAP Mobile Start are authorized to access SAP Document AI. You can control user access in SAP Document AI using default authorization policies via Identity Authentication service. (Learn more)

Configuring tile for SAP Document AI access in mobile

To make SAP Document AI accessible via SAP Mobile Start:

Step 1: Create a New App in Site Manager

In SAP Build Work Zone, navigate to Site Manager > Content Manager and create a new App.

URL: https://<your-tenant-url>/launchpad?sap-ushell-config=headerless#InboundDocument-manageInDocumentAI
ℹ️ Use sap-ushell-config=headerless to hide the launchpad header

Step 2: Configure Navigation Parameters

In the Navigation tab, add the following parameters:

name: Schema name - Optional (from SAP Document AI workspace > Schemas tab)
latest: true, this signifies that only the latest version of the schema is used
client_ID: Client ID (from SAP Document AI workspace > Settings), preset to ‘default’
instance_ID: Instance ID (from SAP Document AI workspace > Settings or BTP account > Instances > Select SAP Document AI instance)
showWorkspaceHeader: true

Step 3: Configure Visualization

You can set your own preferred title, subtitle, icon.

Visualisation Type: Static Tile
Add the following name-value parameters:
mobilestart.type: documentAi
mobilestart.documentAi.clientId: (as specified in Navigation Tab parameters)
mobilestart.documentAi.instanceId: (as specified in Navigation Tab parameters)
mobilestart.documentAi.documentSchemaId: Optional - needed if only a specific schema scenario should be enabled, otherwise all active schemas will be shown to the end user. You can find Schema ID in SAP Document AI workspace > Schemas tab

Screenshot: SAP Document AI Tile Visualization in SAP Build Work Zone

You’ve now created a fully functional SAP Document AI tile for SAP Mobile Start! 🚀

User Experience in SAP Mobile Start

As a business user, you can locate the newly created SAP Document AI tile in the SAP Mobile Start app using the Search function. For easier access, you can also add the tile to your Favorites. When you tap the tile, you’ll be directed to a native document selection screen. Depending on how the service was configured by the administrator, you may have the option to select a document type and schema. If the tile is configured for a specific schema use case, it will be automatically selected.

Screenshot: SAP Document AI tile in SAP Mobile Start App

Upload Options

Scan via Camera – Leverages the device’s native camera with real-time preview, cropping, rotation, and filtering. Multiple scans are compiled into a single PDF.
Select Images from Gallery – Users choose previously captured images which are compiled into a single PDF.
Select File from File Explorer – Users upload a single PDF from their device’s file system.

Screenshot: Scanning a document via device’s native camera

Upload Validation & Handling

Before uploading a document, the app performs several validation checks to ensure compatibility. The file size must not exceed 50 MB, and the document should contain no more than 100 pages. Additionally, file names are automatically sanitized to remove any invalid characters. Once these checks are successfully passed, the upload process begins, with a visible progress bar indicating status. Android users can upload multiple documents synchronously.

After the upload is complete, the document extraction workflow is triggered, and users receive a success notification confirming the process.

Track Status and Review Document

To monitor the progress of document extraction, simply open the Document Overview from the overflow menu within the document selection screen. This action launches the mobile-optimized SAP Document AI UI, providing a clear and intuitive view of the status, allowing reviews and other updates related to your uploaded documents.

Screenshot: Track status and review document via mobile optimized UI

Conclusion

With the release of SAP Mobile Start 2.4 and the seamless integration of SAP Document AI, business users now benefit from a truly mobile-first, intelligent document processing experience. Whether it's capturing receipts on the go, processing invoices, or extracting data from complex forms, SAP Document AI empowers users to automate workflows efficiently—right from their mobile devices. This integration brings the power of AI-driven document handling directly to your fingertips, enabling smarter, faster, and more scalable business operations.

Instantly improve document information extraction

2025-08-28T09:51:15.172000+02:00

Introduction

If you are new to SAP Document AI service, it is a product offering by SAP to provide information extraction from business documents. This primarily helps your business save time by automating and streamlining processes that require information extraction from documents, such as invoices, payment advices, purchase orders, and bills of lading. It limits manual steps, therefore saving time, energy, manpower, and money, which you can invest in growing your business's critical functions.

Instant Learning and Models

As of this writing, and provided you are using Instant Learning, the SAP Document AI uses a multi-model approach to deliver the best possible extraction results. It has mainly three model categories: Pre-trained, LLM, and Rules (Template). Each category may have further different models.

Based on our experiments, it was evident that no single model is capable of achieving higher accuracies than combining them for extraction. This is backed by our evaluation pipeline, which constantly monitors the accuracy of the individual and the combination approach.

We took the facts from the above experiments and designed Document AI naturally using multi-models and optimized extraction based on user inputs on the corrections. This is the foundation of Instant Learning.

How to instantly learn

Start with Schema

A schema is the blueprint of your extraction. End-user defines fields and type of schema. Schema comes in mainly two flavours: Manual and Auto. Manual schemas are limited to template-based extraction, whereas the Auto schemas are more dynamic, powerful, and give the full potential of extraction.

Unless you want to use only the Template model (Rule-based extraction), you should create the Auto schemas, as they can also provide automatic templates depending on when the system thinks that a template could be a good fit for certain layouts.

Now, Auto schemas come in two further flavours depending on the configuration:

Auto with Default-extractor:

This combination works under the hood with all available model types. Pre-trained, LLM and Template.

Auto without Default-extractor:

This combination can automatically provide both LLM and Template extraction without requiring users to specify it.

Recommendation: For the SAP standard schemas, such as Invoice, PO, PA, etc., a user should use "Auto with Default-extractor". For everything else, a user must use the "Auto without Default-extractor".

Extraction, Feedback, and Repeat

Once you have set up your schema and activated it, you are ready for extraction. As you are aware, the following is only applicable to the Auto schema with any flavor.

Your first extraction may not be something you like. It may have errors or inaccuracies. To remove such inaccurate extraction, you need to input it into the system, or in other words, provide feedback to the system.

You can add your feedback in many forms. Let me cover them here:

Feedback as a Schema Description

You are the controller of your extraction. If you want to influence the overall extraction and help the system know your schema better, you can explain that in the description field of the Schema. Your description should resemble a prompt for an LLM.

Feedback as a Schema Field Description

If you see the extraction is inaccurate for a field, then you can add a general instruction for the field, explaining the field, also sharing an example if that is common in all documents for the schema.

Feedback as a Confirm Document

Confirm documents are used to improve models, but without a guarantee. I recommend that you try and test extraction optimization based on your confirmed document-based feedback. Now, confirm that the documents can be used to influence pre-trained model extraction and LLM extraction, depending on your contractual agreement for using SAP Document AI, and allow SAP to use such documents for the extraction.

To confirm a document, the user should carefully review the extraction, correct any incorrect field values, and confirm the document by either making the API call or using the "Confirm" button on the UI.

The Confirm Document not only affects the model but also helps the system select the best model for extraction by switching between models. I will not touch upon the model switching in the blog post, but rather describe how an individual model can be affected with the "Confirm Document".

Pre-trained model

Confirming documents to improve the pre-trained model is a longer cycle and cannot be tested immediately. This requires the engineering teams to scrutinise the data, evaluate training results, and release new models with upcoming releases. This takes time, often weeks.

LLM extraction

Confirm documents can also be used for LLM extraction improvement, which is fundamental to Instant Learning, provided that you opt in for the feature. This optimization is instantly done when a user confirms a document. However, you may still encounter issues with the extraction in rare cases. If the user continues confirming one more document here, the model switching will use the Template model (the rule-based model) to try the extraction further.

Template

Confirm documents are used as samples for automatically created templates. System automatically finds out if you need one more template to optimize your extraction and create it.

After this, the extraction would be at its best possible accuracy. I have omitted the system-level steps for brevity. To learn more about how to do all of these, please visit the SAP Document AI documentation on the official help portal.

Disclaimer: The information in the blog post is my personal opinion and is based on my understanding of the topic. Before making a purchase or using the system in production, please review SAP's official guidance on the product. Please visit the official published help documentation for the product for more information.

Using Document AI Premium in SAP Build Process Automation

2025-09-24T22:46:56.595000+02:00

Intro

SAP Document AI, previously known as Document Information Extraction or DOX, has been bundled in SAP Build Process Automation (SBPA) for many years. The bundled document extraction capabilities in SBPA are leveraging Document AI Base edition based on Machine Learning. Since the introduction of the LLM-based Document AI Premium, there are many cases where the Premium version is the best and prefered option, or you may want to combine both versions. Although SAP Document AI Premium can be connected to SBPA via API like any other BTP service, this still poses a challenge for many SBPA users, who expect a no-code activity similar to the bundled version (DOX SDK).
In this blog post, I will explain how to quickly leverage Document AI Premium in your process automation projects using a reusable package that we deploy in our proof-of-concept activities.

Prerequisites

The assumption is that you have a Document AI Premium service and destination configured on your SBPA subaccount. If not, please check the quick setup tips below and find detailed instructions in SAP Help:

Document AI Premium service: You should setup Document AI Premium in your BTP environment (can be on any subbacount). For CloudFoundry, easiest way is to use booster.
Destination: You should create a destionation, with all required additonal properties for SBPA, in the subbacount where you SAP Build Process Automation is activated and add it to your Build tenant. Easiest way to create a destionation is to download this example and import into your BTP cockpit. You should update 4 parameters (Client ID, Client Secret, Token Service URL and URL) with values from your Document AI Service key.

Easiest way to start

I will first cover the simplest way to get started with Document AI Premium in your automation process, which will work for most users. The goal is to quickly enable the capability to add an activity that will synchronously call your Document AI Premium.

You need to download the latest version of the reusable DocAI Premium package from this link (download the latest version of MTAR file called "(package) Document AI service") and import it into your SAP Build Lobby:

Once the reusable package is uploaded, go to "versions" and select "Publish to Library". Once published, the package will be available to all developers on this tenant.

Open or create an automation project and add this package as a dependency (very similar to the way you add standard SBPA SDK packages):

Now you have the reusable automation 'DocAI - Extract with Schema' in your project, which you can simply drag and drop into your automation flow. You can ignore the other automations for now, as they are meant for advanced cases that I will cover separately. For most users, they are not relevant.

There are 2 input parameters: "fileToUpload" and "schemaName".

"File to upload" is self-explanatory; just provide a local path to a document that you want to process with the Document AI Premium service. "Schema name" refers to any SAP standard or custom schema that is created in your Document AI Premium. Make sure to use the exact name of any standard or custom schema.

And that's it. Once you run your autmation, there will be 3 output parameters that you can use:

fullResponse: Exact full output from the Document AI service (e.g., confidence level, page, coordinates, etc.)
extractedData:Simplified response with only the header and line items fields of your schema. This output is useful if you just want to map extracted data to your Action call, any API or BAPI call, or quickly map exact schema key values.
Status: Status of the extracted document. This field is here for simplification. The status is anyway returned in the fullResponse.status where you can also find additional details if there is a failure. For example, the document may have a valid extension but is corrupted, which will result in a "FAILED" status. To handle this, you can add IF logic after extraction and simply check if the Status == "FAILED" and then take appropriate action.

You can see the difference between the two output responses in the example below:

Notes for advanced automation users

You can double-click on the automation "DocAI - Extract with Schema" and see how this automation works. You will notice that it's uploading the document and waiting for the Document AI Premium service response. You can use the upload automation to send a document to the service, and then start a process that will wait for the service response and continue with the process flow. This asynchronous approach will help you process many more documents with one unattended connection compared to the typical synchronous automation. I will cover this in a separate post, but here is the concept:

You can also activate or deactivate the Instant Learning capability of your Document AI Premium with the automations provided in the package. There is no need to deploy anything. Just add the automations to an empty project and run them in test mode.

For those of you who are familiar with reusable packages (Library) and understand the implications of deploying a new version of your own, here is the editable version of this package so you can adjust it for your specific needs. I don’t advise doing that if you are not familiar with Library concepts and all implications of import and export activities.

Conclusion

I hope this package will speed up and simplify your automation projects with Document AI Premium. Package supports both Windows and Mac OS desktop agents. Let me know how it works for you in the comments below.

I plan to cover the following topics in upcoming posts:

Example automation projects that are leveraging this package
Asynchronous approach for faster processing of large volumes of documents
How to combine new Document AI Workspaces with SAP Build Process Automation

Also, feel free to ask if you want me to cover other topics or if you would like to know more about SAP Document AI Premium combined with SAP Build Process Automation.

Leveraging SAP Document AI to bring intelligent document processing into a classroom

2025-09-30T18:46:43.290000+02:00

Dear SAP University Alliances Community!

We would like to introduce you one of the SAP Business Technology Platform (BTP) services – SAP Document AI – and suggest a scenario how university lecturers can bring it to their classroom.

What is SAP Document AI?

SAP Document AI is an intelligent service that applies advanced machine learning to understand, classify, and process documents with high accuracy. Whether in research, classroom activities, or student projects, it provides a practical way to explore cutting-edge AI applications.

Key Features include:

Document Classification: Automatically categorizes documents by content.
Information Extraction: Accurately identifies and extracts fields, text, and tables.
Deep Learning Models: Ensures reliable performance through advanced algorithms.
Custom Training: Enables adaptation with your own datasets for specialized use cases.

Why SAP Document AI is relevant to academia

SAP Document AI is relevant across a wide range of academic disciplines:

Computer Science & Engineering: Perfect for hands-on projects in AI and machine learning.
Business & Management: Offers insights into automation, process optimization, and information management.
Legal Studies: Simplifies contract review and document analysis.
Healthcare & Life Sciences: Supports efficient processing of patient records and clinical trial data.

By integrating SAP Document AI into teaching and research, educators can bridge the gap between theory and real-world applications.

Bringing SAP Document AI in the educational process

One of the advantages is that SAP Document AI is available through the SAP BTP free trial. This allows students and educators to explore the service directly with no barriers and thus easily experiment with real scenarios, and develop skills that are highly relevant to today’s digital economy.

Essential resources to kickstart mastering SAP Document AI include:

Hands-On Learning Opportunities

The best way to understand SAP Document AI is to try it out. These tutorials and technology blogpost will help you get started step by step in just a little over one hour:

Create a Free SAP BTP Trial Account – suggested duration 15-20 minutes
Use Trial to Set Up Account for SAP Document AI and Go to Application – suggested duration 5-10 minutes
Use Trial to Extract Information from Standard Documents with Generative AI and SAP Document AI – suggested duration 25-30 minutes
Processing a Pokémon Card in 90 Seconds with SAP Document AI

Recommended classroom flow based on the listed above SAP Document AI tutorials:

Session #	Activity	Tutorial to be referred
1	Trial account setup and service activation	Create a Free SAP BTP Trial Account, and then Use Trial to Set Up Account for SAP Document AI and Go to Application
2	Explore the web app, upload sample invoices/POs, and review extraction results with human-in-the-loop validation.	Not relevant
3	Use generative AI to configure extraction for custom document types and then compare outcomes across formats and languages.	Extract Information from Custom Documents
4	Run the “Pokémon Card in 90 Seconds” challenge to demonstrate rapid, end-to-end extraction. Debrief on human-in-the-loop UX and data quality.	Processing a Pokémon Card in 90 Seconds with SAP Document AI
5 (optional)	Connect extraction results to a simple workflow or app (for example, with SAP Build Process Automation, please refer to the respective Sandbox offering from UCC Magdeburg) for an end-to-end process demo.	Not relevant

Ideas for Assignments and Projects

Measure quality of processed documents: Benchmark precision across document templates and languages. Propose prompt or configuration tweaks to improve results.
Design workflows: Build an approval flow that routes low-confidence extractions to human review.
Time-and-cost analysis: Compare manual vs. automated processing time and estimate business value in a mini case study.
Cross-discipline collaboration: Pair business students (process design) with computer science students (API integration) to deliver a working prototype.

Final Thoughts on SAP Document AI

SAP Document AI offers a unique opportunity to combine academic learning with industry-level technology. Whether you’re a student working on your first AI project, or an educator designing courses that reflect the future of work, this tool can make a real difference. Stay tuned for the update in the SAP Universiry Competence Center Magdeburg content on SAP S/4HANA, when a SAP Document AI case will become part of the Global Bike Curricula.

We invite you to explore SAP Document AI, share your experiences, and join the discussion in the comments. Together, let’s bring innovation to the classroom!

Further Inspiration

Looking for ways to bring other SAP innovative technology in the classroom? Explore Academic Lessons by SAP University Alliances. These ready-to-use materials provide practical scenarios to integrate SAP technology into curricula and give students direct exposure to industry-relevant tools.

New Expert-Guided Implementation: Unlocking Joule and SAP Business AI from Strategy to Activation

2025-10-08T00:34:15.690000+02:00

As organizations look to harness the power of Artificial Intelligence (AI) to drive transformation, SAP Business AI stands at the forefront - embedding intelligence directly into business processes to deliver smarter decisions and measurable outcomes.

From finance and supply chain to HR, procurement, and customer experience, SAP Business AI infuses trusted, relevant, and responsible AI across every process - ensuring each insight is contextual, every recommendation actionable, and every outcome aligned with enterprise goals.

Whether leveraging Joule, SAP’s AI copilot, or AI-embedded scenarios across your SAP applications, your path to becoming an intelligent and sustainable enterprise begins with understanding and activating SAP’s Business AI capabilities.

To help you unlock this value faster, we’re launching a new Expert-Guided Implementation (EGI):

Unlocking AI: Joule and SAP Business AI from Strategy to Activation

This multi-day, expert-led workshop is designed to help you and your team unlock the full potential of Joule and SAP Business AI across your SAP landscape. Guided by SAP experts, you’ll explore how to translate strategy into execution with practical exercises, live demonstrations, and activation guidance.

Workshop Schedule

To accommodate global participants, this session will be offered on the following dates and time zones as follows:

2026 Dates	Time	Time Zone
January 19-21	10:00AM - 12:00PM	EMEA \| CET
January 26-28	11:00AM - 1:00PM	NA/LA \| EST
February 16-18	10:00AM-12:00PM	EMEA \| CET
February 23-25	11:00AM - 1:00PM	NA/LA \| EST
March 23-25	10:00AM-12:00PM	EMEA \| CET
March 30-April 1	11:00AM - 1:00PM	NA/LA \| EST

What You’ll Learn

Through this EGI, participants will gain hands-on experience and actionable knowledge in:

SAP Business AI Strategy & Foundation - Understand SAP’s AI vision, architecture, and trusted approach to delivering responsible, embedded intelligence
Commercial Model & AI Units - Learn how to navigate Base vs Premium models, the AI Feature Catalog, and use the AI Estimator to plan your adoption journey
Joule – AI Built for Business - Explore Joule’s unified architecture, roadmap, and capabilities across Joule for Business, Joule for Developers, and Joule for Consultants
S/4HANA AI & ML Capabilities - Experience ISLM in action through AI/ML demo use cases designed to enhance decision-making and process automation
SAP BTP AI & ML Capabilities - Discover how BTP AI Foundation, Document AI, AI Services, AI Core, and AI Launchpad can help you extend and scale innovation
Joule Activation & Prerequisites - Gain a clear view of activation steps, prerequisites, and how to extend Joule with Document Grounding, Joule Studio, and AI Agents

Why Attend

This workshop goes beyond theory—it’s interactive, practical, and guided by SAP experts. By the end, you’ll have:

A clear understanding of SAP’s Business AI strategy and product portfolio
The ability to activate Joule and deploy AI scenarios across your landscape
Actionable steps to integrate AI and Generative AI into your business workflows
Insights into commercial models and AI unit management for smarter planning

Whether you’re a business leader shaping AI strategy or a technical consultant implementing it, this EGI equips you to bridge the gap between strategy and execution.

How to Register

👉 Register here to secure your spot* today to learn from SAP experts and gain the skills to bring Business AI to life across your enterprise.

* You may need to register your S-user for access to SAP Learning Hub to access the EGI registration page. It is a one-time registration, click here.

Joule Prerequisites and Activation: Expert-Guided Implementation

Integrating SAP Document Information Extraction (Document AI) with SAP CAP – Step by Step

2025-10-09T12:50:17.556000+02:00

Introduction :

In the SAP ecosystem, extracting meaningful data from documents such as invoices, contracts, or delivery notes—especially in accounting, finance, and operational processes—is often a manual and time-consuming task.Since these documents are usually in PDF or scanned image formats, using traditional data processing methods to identify fields is both error-prone and unsustainable in the long run.This is where SAP Document AI comes into play.By leveraging AI-powered models, Document AI can automatically analyze documents and extract key fields such as invoice number, date, currency, supplier–customer information, and line items.This eliminates the need for manual data entry, making the overall process faster, more reliable, and significantly more efficient.In this blog post, I will walk you through the step-by-step process of enabling SAP Document AI on SAP BTP and integrating it with a CAP (Cloud Application Programming) project.Starting from service activation, we’ll cover creating a service key, performing authentication (token retrieval), defining actions in CAP, uploading files, and storing the extracted Document AI results in a database.By the end of this guide, you’ll have a fully functional CAP endpoint that can be tested via Postman.You’ll learn how to send real PDF invoice files to Document AI and receive structured data in return—a foundation you can later extend with UI5 interfaces or workflow integrations.

Step 1 — Creating the Document AI Service in SAP BTP

First, we need to enable the Document AI service in SAP BTP. To do this:

Log in to your SAP BTP subaccount.
Open the Service Marketplace .
Find the SAP Document AI Trial service under the Artificial Intelligence category .

On the service details page, you'll see the Overview and Service Plans tabs. Click Create to create a service instance .

Select the default plan . This plan enables basic API access in a trial environment and creates the instance your CAP project needs to connect to the Document AI service.

Fill out the form below:

Service: SAP Document AI Trial Version
→ We will be using this service. Since it is a trial version, we will use the free trial quota.
Plan: default
→ Only the default plan is available in the trial environment. It is sufficient for basic use.
Runtime Environment: Cloud Foundry
→ Defines where the service will run. Cloud Foundry is often chosen for integration with CAP projects.
Domain: dev
→ Select the Cloud Foundry domain where your project is located (for example, dev, test).

Instance Name: DocAI
→ You can give the service instance any name you want. This makes it easier to connect later (you can use lowercase, for example docai).

After filling in all the fields, you can click the Next button to proceed to the confirmation screens and complete the sample creation process.

Once the instance has been successfully created, the next step is to generate a Service Key for it . This key contains the authentication information needed for your CAP project to connect to the Document AI service.

Click the ellipsis ( … ) on the far right of your sample row .
Select Create Service Key from the menu .

In the dialog box:

- In the Name field, enter a meaningful name such as DocAIKey .
- You can leave the other settings at their default values.
- Then click Create .
Once the Service Key is generated, click on its row to open the detail page.

After copying the JSON:

Adding Service Key details to the file To access .envthe Document AI service, we need to define credentials in our CAP project. To do this,
you can add a structure like the following to the file located in your project's root directory:.env

Correct the highlighted errors and try again.

Unsupported HTML code has been removed from your message. The HTML has been updated. Please review the message and send it when you are satisfied.

{
  "DOCAI_CLIENT_ID": "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx",
  "DOCAI_CLIENT_SECRET": "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx",
  "DOCAI_TOKEN_URL": "https://39c50baetrial.authentication.us10.hana.ondemand.com/oauth/token",
  "DOCAI_URL": "https://aiservices-trial-dox.cfapps.us10.hana.ondemand.com/document-information-extraction/v1"
}

In real projects , instead of manually adding these details to the file .env, service keys are automatically linked to the application. This way, the application reads credentials from environment variables ( ) and is managed more securely. Automatic linking can also be achieved during deployment via MTA or manifest files VCAP_SERVICES.

service.cdsAction Definition in File

using { app as db } from '../db/schema';

service InvoiceService {
  entity Invoices as projection on db.Invoices;
  entity InvoiceItems as projection on db.InvoiceItems;

  action analyzeAndSave(file : LargeBinary, filename : String) returns {
     documentNumber : String(60);    
    documentDate   : Date;           
    currencyCode   : String(3);      
    senderName     : String(255);    
    receiverName   : String(255);   
    netAmount      : Decimal(15,2);  
    taxName        : String(40);     
    taxRate        : Decimal(5,2);   
    taxAmount      : Decimal(15,2);  
    grossAmount    : Decimal(15,2);  

  };
}

First, we define a CAP Action to send the PDF to the Document AI service and retrieve the result . This action can be triggered from the frontend or from tools like Postman.

For example, let's define an action with the following name :analyzeAndSave

filefield: Retrieves PDF file content in Base64 format.
filenamefield: Contains the file name.
returnssection: Returns key invoice fields from the AI service (for example, documentNumber, documentDate, currencyCode, etc.).

This step creates the backend point that will trigger access to the Document AI service.

At this point, we need to listen for the action we defined in the backend . To do this , we add an event handler to . When the action is triggered, this handler sends the file to the Document AI service, retrieves the analysis result, and saves it to the database. The entire flow is managed within this function.analyzeAndSaveservice.cdssrv/service.js

module.exports = cds.service.impl(function () {
  this.on("analyzeAndSave", async (req) => {
    // We define the **helper functions required for the Document AI integration**.

  });
});

Obtaining a Token (Authentication)
Now that we've defined the service credentials in the file.env, we need to obtain an access token to send requests to the Document AI service.We'll add the following sample functionto our file:
getTokensrv/service.js

const axios = require("axios");

async function getToken() {
  const r = await axios.post(
    process.env.DOCAI_TOKEN_URL,
    new URLSearchParams({
      grant_type: "client_credentials",
      client_id: process.env.DOCAI_CLIENT_ID,
      client_secret: process.env.DOCAI_CLIENT_SECRET,
    }),
    { headers: { "Content-Type": "application/x-www-form-urlencoded" } }
  );
  return r.data.access_token;
}

This function access_tokenreturns a value from the SAP Document AI service using the information in the file. This token will be used in subsequent API calls within the Authorization header ..env

In real projects, this step is often automated through target+service binding . The CAP application reads the necessary information VCAP_SERVICESfrom environment variables to obtain the token .

In this step, we retrieve the schemas defined in Document AI and
select the appropriate one for invoices. This allows the model to determine which fields to extract (e.g., Invoice ID, Date, Currency, etc.).

const schemas = await fetchSchemas(base, token);
console.log("[DocAI] schemas count =", schemas.length);

const chosen = pickInvoiceSchema(schemas);
if (!chosen) return req.error(500, "No invoice-related schema available in DOX");
console.log("[DocAI] chosen =", chosen);

- fetchSchemas: Retrieves the schema list from DOX using the access token.
- pickInvoiceSchema: Selects the schema whose name is invoice or contains the invoice .documentType
- If none are found, return 500 (an obvious error to the developer).
Base64 → Buffer conversion (prepare the file)
The PDF file received from the client (UI5/Postman) may be in Base64 format. If it is a Data URL (for example,data&colon;application/pdf;base64,...), remove the prefix and convert it to a true binary buffer.

let b64 = Buffer.isBuffer(file) ? null : String(file);
if (b64 && b64.startsWith("data:")) {
  const i = b64.indexOf("base64,");
  if (i > -1) b64 = b64.slice(i + "base64,".length);
}
const buffer = Buffer.isBuffer(file) ? file : Buffer.from(b64 || file, "base64");
console.log("[CHECK] filename=", filename, "buffer length=", buffer?.length);
if (!buffer?.length) return req.error(400, "Empty or invalid base64 file");

- The field we will send via HTTP form-data needs to be binary ; therefore we convert the Base64 payload into a buffer .file
- We return 400 (client error) for an empty or corrupt file .
Create a job (submit with schemaId) We provide the selected schema and document type in , then submit the file to DocAI to start.
options

const options = {
  clientId: "default",
  schemaId: chosen.id,
  documentType: chosen.documentType || "invoice",
};
console.log("[DocAI] submit options =", options);

const submitted = await submitJob(base, token, buffer, filename || "invoice.pdf", options);
const jobId = submitted?.id;
if (!jobId) throw new Error(`Job submit failed: ${JSON.stringify(submitted)}`);
console.log("[DocAI] jobId =", jobId);

- schemaId: chosen.id→ we use a dynamic schema id.
- If successful, it returns a jobId; we use this ID to track the status of the job.
- If filenamenot provided the default value is "invoice.pdf".
Question the work and get results
- The job runs in the background. We poll at intervals until the status is COMPLETED .
- Once the process is complete, the extracted areas become available under .extraction

const result = await pollJob(base, token, jobId, 60000, 1500);
console.log("[DocAI] status DONE, extraction keys =", Object.keys(result?.extraction || {}));

- pollJob(base, token, jobId, 60000, 1500)→ 60 sec timeout / 1.5 sec interval .
- Below and result.extractionyou get .headerFieldslineItems
NOW LET'S GO TO TESTING:
First, ...

cds watch --profile hybrid

Let's start our CAP project with the following command. (I started it this way because I chose HANA as the database .)

Sample Invoice PDF file:

Above is a sample PDF file we sent to the SAP Document AI service. The service automatically analyzes the invoice and extracts fields such as the invoice number, date, currency, supplier-customer information, and item information.

Testing with Postman:
Now that everything is ready on the backend, we can test the CAP action integrated with the Document AI service. To do this, use a REST client like Postman and send the following request:

Method: POST
URL: http://localhost:4004/odata/v4/invoice/analyzeAndSave
Body (JSON):
- file name: Name of the PDF file to be uploaded
- file: PDF file content in Base64 format
In this answer:
- The fields extracted by Document AI match exactly the action return types defined in service.cds.
- These values are also automatically saved to the HANA database in the background.
In real projects, this setup isn't written manually. Instead, the Document AI service automatically connects to the CAP application via service binding . The application reads the necessary credentials directly from environment variables, and token acquisition is performed automatically in the background. This not only increases security (sensitive data like client secrets aren't visible in the code or repository) but also allows the application to run across different environments (dev, test, production) without additional configuration . During deployment, binding is provided via MTA or manifest files, and the application continues to run seamlessly with the same approach when deployed to production..envVCAP_SERVICES

Optional: Storing Raw Document AI JSON for Debugging:

In the Invoices entity of the schema , I defined a field called to store the original response returned by the Document AI service. This field is of type LargeString and holds all the header and line item data returned by Document AI in JSON format.RawJson

The purpose of this area:

Debug/Trace : Easily inspect the data returned by Document AI during development.
Extensibility : If I later want to add new fields or change the mapping structure, I can access the original source.
Data Traceability : Store raw invoice data so it can be reprocessed when necessary.

For example, when an invoice is processed, all the extraction data returned by Document AI is stored in this column. You can then open this data and clearly see which fields Document AI has recognized.

In this article, we explored how to enable SAP Document AI on SAP BTP and integrate it end-to-end with a CAP project. Each step—from creating the service instance and generating the service key to authenticating, defining CAP actions, and processing documents—represents a fundamental building block of a real-world backend automation setup.With this integration, you can now automatically process invoices or similar business documents without manual data entry. The extracted fields can be stored in your HANA database and then used for reporting, process automation, or analytics. This approach not only speeds up financial and operational workflows but also reduces human error and increases overall process reliability.In production environments, this setup is typically secured and automated through service binding, allowing applications to read credentials from environment variables without manual configuration. This ensures smooth operation across development, test, and production landscapes.As a next step, you can extend this backend integration with a SAP UI5 dashboard or management panel, enabling users to upload, view, validate, and approve invoices in real time. By doing so, you’ll move from a simple technical integration to a fully functional end-to-end intelligent document processing solution.

New Expert-Guided Implementation: Unlocking Joule and SAP Business AI from Strategy to Activation

2025-10-13T14:51:50.787000+02:00

To help you unlock this value faster, we’re launching a new Expert-Guided Implementation (EGI):

Unlocking AI: Joule and SAP Business AI from Strategy to Activation

Workshop Schedule

To accommodate global participants, this session will be offered in two time zones:

October 20–22, 2025 | 10:00am–12:00pm EMEA time zone
November 3–5, 2025 | 10:00am –12:00pm NY time zone

What You’ll Learn

Through this EGI, participants will gain hands-on experience and actionable knowledge in:

SAP Business AI Strategy & Foundation - Understand SAP’s AI vision, architecture, and trusted approach to delivering responsible, embedded intelligence
Commercial Model & AI Units - Learn how to navigate Base vs Premium models, the AI Feature Catalog, and use the AI Estimator to plan your adoption journey
Joule – AI Built for Business - Explore Joule’s unified architecture, roadmap, and capabilities across Joule for Business, Joule for Developers, and Joule for Consultants
S/4HANA AI & ML Capabilities - Experience ISLM in action through AI/ML demo use cases designed to enhance decision-making and process automation
SAP BTP AI & ML Capabilities - Discover how BTP AI Foundation, Document AI, AI Services, AI Core, and AI Launchpad can help you extend and scale innovation
Joule Activation & Prerequisites - Gain a clear view of activation steps, prerequisites, and how to extend Joule with Document Grounding, Joule Studio, and AI Agents

Why Attend

This workshop goes beyond theory—it’s interactive, practical, and guided by SAP experts. By the end, you’ll have:

A clear understanding of SAP’s Business AI strategy and product portfolio
The ability to activate Joule and deploy AI scenarios across your landscape
Actionable steps to integrate AI and Generative AI into your business workflows
Insights into commercial models and AI unit management for smarter planning

Whether you’re a business leader shaping AI strategy or a technical consultant implementing it, this EGI equips you to bridge the gap between strategy and execution.

How to Register

Register here to secure your spot* today to learn from SAP experts and gain the skills to bring Business AI to life across your enterprise.

* You may need to register your S-user for access to SAP Learning Hub to access the EGI registration page. It is a one-time registration, click here.

[2025-Q4] SAP BTP Adoption Webinars – Unlock the Power of AI, Data, and Integration

2025-10-23T20:29:52.412000+02:00

Get ready for a packed lineup of expert-led sessions designed to help you accelerate innovation with SAP Business Technology Platform (BTP). From AI best practices to integration and analytics, these webinars will equip you with practical insights and real-world examples you can apply right away.

📅 Oct 30 | Pipeline Concept in SAP Integration Suite

👉Learn how the pipeline concept accelerates your migration from SAP Process Orchestration to SAP Integration Suite.

🔗 Register Now

📅Nov 04 | Service and Service Excellence with AI Driven Insights

👉Experience how SAP solutions transform intelligence into action—turning service into a seamless, revenue-driving experience.

🔗 Register Now

📅 Nov 05 | Best practices for secure and scalable LLM model usage in BTP AI apps with SAP AI Core

👉 Discover how to make secure and scalable large language model usage in BTP AI applications with SAP BTP AI Core.

🔗 Register Now: EMEA/MEE/APAC Session | Americas/EMEA Session

📅 Nov 13 | AI-Driven Business Solutions – Exploring the Potential of SAP BTP AI - Americas

👉 Discover how enterprises are turning AI hype into real business outcomes with SAP BTP.

🔗 Register Now

📅 Nov 13 | Enablement on SAP Mobile Start and SAP Build Work Zone with Showcase of Joule as AI Copilot

👉 Discover how to leverage SAP Mobile Start and SAP Build Work Zone as central entry point with Joule AI Copilot for connected SAP solutions.

🔗 Register Now

📅 Nov 19 | AI-Driven Business Solutions – Exploring the Potential of SAP BTP AI - APAC/EMEA

👉 Discover how enterprises are turning AI hype into real business outcomes with SAP BTP.

🔗 Register Now

📅 Nov 20 | BTP AI Best Practices – Knowledge Graph-based Retrieval Augmented Generation

👉 Learn how to combine Knowledge Graphs with Large Language Models for advanced RAG scenarios.

🔗 Register Now: EMEA/MEE/APAC Session | Americas/EMEA Session

📅 Dec 03 | BTP AI Best Practices – Vector-based RAG, Document Grounding, and Examples

👉 See how to ground large language models on company-specific data using SAP Document Grounding.

🔗 Register Now: EMEA/MEE/APAC Session | Americas/EMEA Session

📅 Dec 09 | AI-Driven Business Solutions – Exploring the Potential of SAP BTP AI (Americas Session)

👉 Discover how enterprises are turning AI hype into real business outcomes with SAP BTP.

🔗 Register Now

📅 Dec 15 | AI-Driven Business Solutions – Exploring the Potential of SAP BTP AI (APAC/EMEA Session)

👉 Discover how enterprises are turning AI hype into real business outcomes with SAP BTP.

🔗 Register Now

💡Don’t miss the opportunity to learn from SAP experts and accelerate your journey to the Intelligent Enterprise.

SAP Document AI in SAP MDG using Intelligent Scenario Lifecycle Management (ISLM)

2025-10-29T05:10:35.628000+01:00

Introduction

In this blog, let us understand how to leverage SAP Intelligent Scenario Lifecycle Management (ISLM) to manage Generative AI and AI business services in the SAP MDG. This blog mainly focuses on how to use one of the AI Business Services i.e. Document AI in MDG via ISLM. It is formerly known as Document Information Extraction (DIE).

MDG Scenario

The scenario is to derive the address details from vendor registration PDF document which will be uploaded in the MDG change request. SAP Document AI from BTP is used to read the details from a PDF document and derive the data into MDG change request.

Implementation

1. BTP

First of all, The BTP account is needed to create instance on the AI business service - SAP Document AI to read data from PDF. If you do not have a BTP account, please refer this link to create a BTP trial account Get an Account on SAP BTP Trial | SAP Tutorials.

You can find the SAP Document AI Trail (DIE) in the service marketplace to create an instance.

Please make sure instance as well as service key is created. The service key is used later in the ISLM connection mapping configuration.

Please subscribe to SAP Document AI UI application

Assign Role Collection to your user in Security.

To simplify all the above setup, you can use Booster as alternative to setup the SAP Document AI in BTP.

PDF Document

Please find sample PDF document which will be used as a document template to setup in the SAP Document AI.

Schema Configuration

Create a custom schema in SAP Document AI Application to extract fields from PDF document.

Add Header Fields

Add header fields in the custom schema to map with PDF document. All the fields such as FirstName, LastName, House No, Street, postal code etc. from PDF to map with business partner fields in MDG change request.

Activate the schema after all the fields are added.

Create Template

Template in SAP Business AI that uses schema and associated document. You could edit the extraction results of the associated documents to teach SAP Document AI where certain field values are located on the document to improve the accuracy of the extraction.

There is no associated document, so we need to add PDF document now in Documents.

Add Document

Add the PDF document in the Documents in SAP Document AI application and map the schema fields with PDF fields.

Add this document to template

Now, click on Add to Template Button to add this document to custom template created in one of the above steps.

Activate template

Access Template on the left navigation pane to activate it.

Now, all the setup related to SAP Document AI is completed.

2. Intelligent Scenario Lifecycle Management (ISLM) in MDG System

ISLM is part of the ABAP Platform offered as core functionality in ABAP stack.
ISLM Framework is developed in SAP BASIS Component from ABAP Platform 2020 delivery with SAP BASIS 755 / 7.8x onwards.
ISLM facilitates and helps to consume the Machine learning scenarios with an ABAP based Business Application seamlessly
ISLM helps to integrate AI and AI Business services into MDG system (S/4HANA).
ISLM offers an end-to-end framework to manage both pre-delivered and custom AI scenarios in SAP S/4HANA.

In this example, would like to walk through custom AI-DAR scenario through ISLM.

The below configuration is to be completed in MDG system to enable ISLM related applications.

Please make sure that role SAP_BR_ANALYTICS_SPECIALIST is assigned to the user which is required to access the below Intelligent Scenario Apps.

Click on Intelligent Scenarios App to create an intelligent scenario to consume AI business service SAP Document AI (DIE).

As you may aware, SAP ISLM provides two options as Embedded and Side-by-Side. In this example, would like to cover side-by-side scenario. So, click on create Button to initiate a side-by-side scenario creation.

Enter Scenario Name, Description and Type as Document Information Extraction.

It is recommended to use $TMP local package in ISLM until the Intelligent scenario is developed and tested thoroughly. After finalization, create the application package specific intelligent scenario. ISLM creates 'Intelligent Scenario' as an ABAP Object (Transportable / added to a TR based on catalog/package).

Create a prediction class with interface IF_ISLM_INTS_SBS_DOCX which will be assigned in the intelligent scenario.

The intelligent scenario could be published after all the above steps are done.

Maintain connection mapping between intelligent scenario and SAP Document AI (DIE) service (BTP) using service key and oAuth profile through the transaction ISLM_CONN_MAP.

In the Intelligent scenario management App, the deployment of intelligent scenario needs to be activated.

3. Master Data Governance (MDG)

In MDG, Cross entity derivation BADI could be used to call the Document AI via ISLM in MDG change request. In the BADI, it will read the name and address details using Document AI from PDF, which is attached in the MDG CR, and it will derive this information in the CR.

Execution

To execute this scenario, attach the PDF document in the MDG CR header and select BP grouping in the CR, it will trigger the BADI to derive the name and address details from PDF to MDG change request.

The file name and file title could be any and it is not fixed to any particular name.

Now, the name and address details are derived in the CR from PDF document.

Conclusion

In this blog post, we have learnt how to derive fields in the MDG change request from PDF document using SAP Document AI via ISLM.

Joule - Document Grounding with SAP Build Work Zone, advanced edition (RAG)

2025-11-07T01:08:40.886000+01:00

Setting up Document Grounding for SAP Build Work Zone, advanced edition

Document grounding (SAP’s implementation of RAG) is a foundational Joule capability which can provide more comprehensive responses by drawing from business documents located in SAP and third-party repositories.

Grounding is a service designed to handle data-related tasks, such as grounding and retrieval, using vector databases. It provides specialized data retrieval through these databases, grounding the retrieval process with your own external and context-relevant data. Grounding combines generative AI capabilities with the ability to use real-time, precise data to improve decision-making and business operations for specific AI-driven business solutions.

Grounding converts user provided documents into vector representations which are stored as a database. The indexing pipeline preprocesses unstructured and semi structured data into chunks and embeddings and the retrieval pipeline takes incoming user queries and converts them into vector representations. The query vectors are used to search the database and retrieval relevant information.

Prerequisites: Before triggering the setup process, make sure that you’ve purchased the AI Unit SKU (8019164) OR AI Units SKU - 8018592. After you’ve done so, the entitlements for document grounding are automatically added to your global account. Validate if you can see the entitlement "Document Grounding" in Global Account -> Service Assignment area.

Document Grounding now supports two authentication mechanisms. Please check the Configure Access from SAP BTP before you start your setup.

This blog focuses on setting up Document Grounding with SAP Build Work Zone, advanced edition. If you need to setup Document Grounding for Microsoft Sharepoint, please refer to this blog written by Nagesh: https://community.sap.com/t5/technology-blog-posts-by-sap/joule-getting-started-with-document-grounding-setup-guide/ba-p/13739501

Document types supported for SAP Build Work Zone, advanced edition: Blog posts, DOCX, JPEG, JPG, JSON, Knowledge base articles, PDF, PNG, PPTX, TIFF, TXT, Wiki pages (https://help.sap.com/docs/joule/integrating-joule-with-sap/set-up-document-grounding)

Setup Steps

Step 1: Enable Document Grounding capabilities in SAP Build Work Zone, advanced Edition

Step 2: Enable Document Grounding for Workspaces & Sub-Workspaces in SAP Build Work Zone, advanced Edition

Step 3: Explore Content Locations — Blogs, Wikis, and Documents for Grounding

Step 4: Use an Existing Joule Subaccount or Create a New One to Set Up Document Grounding

Step 5: Create Certificate and RSA Key Files

Step 6: Run cURL commands to setup Document Grounding pipelines

Step 7: Verify and Test Joule

Step 8: Appendix

Step 1: Enable Document Grounding capabilities in SAP Build Work Zone, advanced Edition

Navigate to SAP Build Work Zone, advanced Edition → App Finder → Administration Console.

In the Administration Console, go to External Integrations → OAuth Clients, and click Add OAuth Client.

Enter a Name and an Integration URL for the OAuth client.

Note: The Integration URL is not validated, so you can enter any placeholder value.

After saving the OAuth Client, click View to display the Client Key and Client Secret.
Copy these values. You will need them later during the integration setup.

OAuth Client Key: dgkey123456

OAuth Client Secret: dgsecretabcdef

Navigate to Administration Console → Feature Enablement → Features
Enable the Document Grounding Integration feature, and then select the OAuth Client you created in the previous step.

From the Administration Console → Overview page, copy the DWS URL.
For example: https://abcdef.eu10-1.dws.workzone.ondemand.com

Step 2: Enable Document Grounding for Workspaces & Sub-Workspaces in SAP Build Work Zone, advanced Edition

Workspaces and Sub-Workspaces both need to be enabled for Document Grounding. Selecting only the "Workspace" does not automatically enable Document Grounding in "Sub-Workspaces"

Navigate to SAP Build Work Zone, advanced Edition → Workspaces → New Workspace, and select Public Workspace.

Note: Document Grounding is supported only for Public Workspaces. If you have an existing Private Workspace, you can convert it to a Public Workspace from the Workspace Admin Settings.

In the workspace, navigate to Workspace Admin Settings → Edit Workspace.

Select the checkbox for Document Grounding to enable grounding for the workspace and then click Save

Note: If your setup includes a Sub-Workspace, make sure to enable Document Grounding there as well.
Enabling it at the main Workspace level does not automatically activate it for any Sub-Workspaces.

You can also manage Document Grounding across all workspaces by navigating to SAP Build Work Zone, advanced Edition → Administration Console → Workspaces.

Step 3: Explore Content Locations — Blogs, Wikis, and Documents for Grounding

Documents in SAP Build Work Zone, advanced edition can be uploaded as individual files or referenced from other content areas such as Blogs, Wikis, and Knowledge Base articles.
In these cases, the entire page or article is treated as grounding content. All such sources are indexed and leveraged by Joule to provide accurate, context-aware responses based on your organization’s information.

3.a Content / Files:
In your workspace, go to Content to view existing files or upload new documents that will be used for document grounding.

3.b Knowledge Base:
In your workspace, go to Knowledge Base to create and manage knowledge base articles that can be used for document grounding.

3.c Blog or Wiki:
In the Content area, click Create to add a new Blog or Wiki page. These pages will be indexed and used as grounding content.

Step 4: Use an Existing Joule Subaccount or Create a New One to Set Up Document Grounding

Document Grounding must be configured in the same subaccount where Joule (or Unified Joule) is enabled.
Select the subaccount where Joule is already set up and verify that it is hosted in a supported region.

AWS: ap10 – Australia (Sydney), ap11 – Asia Pacific (Singapore), eu10 – Europe (Frankfurt), eu11 – Europe (Frankfurt, EU Access), jp10 – Japan (Tokyo), us10 – US East (VA)
Google Cloud: ap30 – Australia (Sydney), eu30 – Europe (Frankfurt), us30 – US Central (IA)
Microsoft Azure: ap20 – Australia (Sydney), eu20 – Europe (Netherlands), us21 – US East (VA)

For the latest list of supported regions, refer to the official documentation: Supported Regions for Document Grounding

4.a Scenarios for Setting Up Document Grounding

Depending on how SAP Build Work Zone is configured in relation to your Joule formation, there are four possible setup paths. As mentioned earlier, the Document Grounding service must always be added to the Joule subaccount.
However, depending on your environment, SAP Build Work Zone, advanced edition may be deployed in the same subaccount or in a different one.
The following options outline the supported configurations.

4.a.1 – Unified Joule Already Exists with SAP Build Work Zone, standard Edition

You already have Joule configured, and SAP Build Work Zone, standard Edition is part of the same Joule formation.
In this case:

Joule remains integrated with SAP Build Work Zone, standard Edition (the Joule icon will only appear there).
You can still use SAP Build Work Zone, advanced edition for Document Grounding, but it won’t display the Joule icon because only one system instance can be part of a Joule formation.

Steps

Navigate to your Joule subaccount in BTP Cockpit.
Go to Entitlements → Add.
Search for and add the “Document Grounding” service plan.

4.a.2 – Unified Joule Already Exists with SAP Build Work Zone, advanced edition

You already have Joule set up and integrated with SAP Build Work Zone, advanced Edition.
In this scenario, you can directly enable Document Grounding in the same subaccount.

Steps

Navigate to your Joule subaccount.
Go to Entitlements → Add.
Search for and add the “Document Grounding” service plan.

Reference (optional): For more details on Setting up Unified Joule for scenarios 4.a.3 and 4.a.4 below you can refer to the SAP Community Blog:
Joule End-to-End Setup Guide for All Lines of Business – Unified Approach

4.a.3 – Setting Up Unified Joule for the First Time and Connecting an Existing SAP Build Work Zone, advanced edition

You are enabling Unified Joule for the first time and want to use an existing SAP Build Work Zone, advanced Edition that resides in a different subaccount.
In this setup, the Joule subaccount will host Joule and Document Grounding services only—no Work Zone Standard—while your existing Work Zone Advanced Edition will be linked to Joule during the booster configuration.

Steps

Create a new subaccount in your global account to host Joule.
Navigate to Trust Configuration → Establish Trust and set up trust with Cloud Identity Services (CIS / IAS).
Go to Entitlements → Edit → Add Service Plans, and add:

Joule
Document Grounding

From your Global Account, open Boosters → Set Up Joule, and click Start.
When prompted, select your new Joule subaccount and proceed through the booster steps.
In the booster flow, choose to connect your existing SAP Build Work Zone Advanced Edition tenant to the Joule formation.
Complete the booster to finish the configuration. Once successful, Joule will be active in the Joule subaccount and integrated with your existing Work Zone Advanced Edition tenant.

4.a.4 – Setting Up Unified Joule for the First Time (SAP Build Work Zone, advanced edition in the Same Joule Subaccount)

You are setting up Unified Joule for the first time and intend to deploy SAP Build Work Zone, advanced Edition directly within the same Joule subaccount (rather than linking an existing external Work Zone tenant).
This configuration keeps all related services — Joule, Document Grounding, and Work Zone Advanced — together in one managed environment.

Steps

Create a new subaccount in your BTP Cockpit to host Joule and Work Zone Advanced.
Navigate to Trust Configuration → Establish Trust, and set up trust with Cloud Identity Services (CIS / IAS).
Go to Entitlements → Edit → Add Service Plans, and add the following:

Joule
SAP Build Work Zone – Advanced Edition
(add both advanced and advanced (application) service plans)
Document Grounding

From your Global Account, open Boosters → Get Started with SAP Build Work Zone, advanced Edition, and complete the booster to provision the Work Zone Advanced tenant.
Refer to the official help documentation for detailed guidance:
Onboarding to SAP Build Work Zone Advanced Edition
After Work Zone Advanced is successfully provisioned, open Boosters → Set Up Joule, and click Start.
Select your newly created subaccount and proceed through the Joule booster steps.
When prompted, choose SAP Build Work Zone, advanced Edition to integrate it with Joule.

Optional Steps

If you would like to enable Joule in SAP Build Work Zone, advanced edition, refer to the following blog for detailed guidance:
🔗 Joule is Now Integrated into SAP Build Work Zone, Advanced Edition

If you would like to enable Joule in SAP Build Work Zone, standard edition, follow the steps described in this blog:
🔗 Activate Joule with SAP Build Work Zone and SAP Mobile Start

After completion of Step 4.a.1 or 4.a.2 or 4.a.3 or 4.a.4 proceed to create a Destination

4.b Create Destination

In your Joule Subaccount, navigate to Destinations → Create → From Scratch. Enter a Name for the destination. Add the values provided below

Name: HGDGBWZ924 (You can have any name)

Property	Value
URL	The URL is https://<DWS URL>/api/v1/dg-pipeline/metadata Check DWS URL in the SAP Build Work Zone, advanced edition -> Admin Console -> Overview E.g: https://abcdef.eu10-1.dws.workzone.ondemand.com/api/v1/dg-pipeline/metadata
Type	HTTP
Proxy Type	Internet
Authentication	OAuth2ClientCredentials
Client ID	Enter the value of the OAuth client Key from BWZ Advanced created in Step 1 OAuth Client Key: dgkey123456
Client Secret	Enter the value of the OAuth client Secret from BWZ Advanced created in Step 1 OAuth Client Secret: dgsecretabcdef
Token Service URL Type	Dedicated
Token Service URL	Enter your <DWS URL>/api/v1/auth/token E.g: https://abcdef.eu10-1.dws.workzone.ondemand.com/api/v1/auth/token Check DWS URL in the BWZ Advanced -> Admin Console -> Overview
Additional Properties	(Click New Property and manually add these properties to the destination)
HTML5.DynamicDestination	true
SetXForwardedHeaders	false
HTML5.SetXForwardedHeaders	false

4.c Create Document Grounding Service Instance

In your Joule Subaccount, go to Service Marketplace → Document Grounding → Create.

Select Runtime Environment: Other
Enter an Instance Name: for example, groundingcli
Click Create

Create a Service Binding for the Document Grounding Instance

In your Subaccount, navigate to Instances & Subscriptions → Instances
Select your Document Grounding Service Instance, then proceed to "Create Service Binding"

Enter a Binding Name, for example: groundingkey, and click Create

Copy the mTLS Document Grounding Service Binding URL (MTLS URL) for later use.

E.g: https://mtls.rage.c-1111.kyma.ondemand.com

4.d Create Cloud Identity Services Instance

In your Subaccount, navigate to Services → Service Marketplace → Cloud Identity Services → Create.

Plan: application
Runtime Environment: Other
Instance Name: groundingCIS

Then click Next (do not click Create yet).

In the Parameters section, enter the following JSON — replacing <doc-grounding-instance-name> with the Document Grounding Service Instance Name you created earlier (E.g: groundingcli) — and then click Create.

{
   "consumed-services":[
      {
         "service-instance-name":"<doc-grounding-instance-name>"
      }
   ]
}

In your Subaccount, navigate to Instances & Subscriptions → Instances
Select your Cloud Identity Services Instance, then proceed to "Create Service Binding"

Enter a Service Binding Name, for example: cisSK, and provide the following parameters. Then click Create

{
    "credential-type": "X509_GENERATED",
    "validity": 365,
    "validity-type": "DAYS"
}

Download the Service Binding file for the Cloud Identity Services instance.

Step 5: Create Certificate and RSA Key Files

You’ll now create two files required for secure communication between Document Grounding and Cloud Identity Services (CIS).

Open the CIS Service Binding file and copy the following values:

clientid
authorization_endpoint

Then, update the authorization_endpoint by replacing /oauth2/authorize with /oauth2/token

E.g: clientid: aaa-bbb-ccc-ddd-eee

E.g: authorization_endpoint: https://abc2wxyz.accounts.ondemand.com/oauth2/token

Copy the certificates (there are 2 certificates) from the CIS Service Binding file (Starting with "----BEGIN CERTIFICATE---- to ----End CERTIFICATE----" and RSA key (Starting with ----BEGIN RSA PRIVATE KEY---- to ----END RSA PRIVATE KEY----) to VSCode (or Notepad++)

In VSCode press Ctrl+H and enter "\n" in Find and "," in Replace. Select "Match Case" and click "Replace all"

After that enter "," in Find and "\n" in Replace. Select "Match Case" and "Use Regular Expression" and click "Replace all"

You will see "2 Certificates" and "1 RSA Key" similar to the screenshot below

Save the two certificates into a single file and save it as a .crt file.
Example: DocumentGrounding.crt
Save the RSA private key as a .key file.
Example: DocumentGrounding.key

Step 6: Run cURL commands to setup Document Grounding pipelines

In this step, you’ll use Bruno (or any API client such as Postman) to execute cURL commands that create and configure Document Grounding pipelines in your Joule subaccount.
These pipelines establish the connection between SAP Build Work Zone, advanced edition content and the Document Grounding service.

6.a Install Bruno and Create a Collection

Install Bruno: Download and install the Bruno API Client.

Open Bruno and click Create Collection

Provide a Name for the collection, for example: DGcollection, choose a Location to save it, and click Create.

Before proceeding, ensure you have the following two URLs ready:

IAS URL (of Joule Subaccount): https://abc2wxyz.accounts.ondemand.com
Document Grounding Service Binding URL (mTLS): mtls.rage.c-1111.kyma.ondemand.com

In Bruno, perform the following steps:

Select your collection (DGcollection).
Go to Client Certificates.
For the Domain, enter your IAS URL — abc2wxyz.accounts.ondemand.com
Select the Certificate File (DocumentGrounding.crt) and the Key File (DocumentGrounding.key).
Click Add to save the configuration.

Repeat the same process for the Document Grounding Service Binding URL:

Domain: mtls.rage.c-1111.kyma.ondemand.com
Certificate File: DocumentGrounding.crt
Key File: DocumentGrounding.key

Click Add to save the configuration.

6.b Create a Bearer Token

In this step, you’ll generate an OAuth 2.0 bearer token to authenticate subsequent API calls for creating Document Grounding pipelines.

In Bruno, select your collection (DGcollection).
Click New Request and enter:
- Name: Get Bearer Token
- URL: Your CIS URL followed by /oauth2/token
  Example: https://abc2wxyz.accounts.ondemand.com/oauth2/token
Click Create.

Add Parameters

client_id: (from your Cloud Identity Services binding file cisSK.json) Example: aaa-bbb-ccc-ddd-eee
grant_type: client_credentials

Add Headers

content-type: application/x-www-form-urlencoded
accept: application/json

Change the method from GET to POST, click Save, then click the → (Send) button.

You will receive a response similar to the example below.
Copy the access_token, as it will be used in the next step to create the Document Grounding Pipeline

{
  "access_token": "xyzxyzxyzxyzxyzxyzxyz",
  "token_type": "Bearer",
  "expires_in": 3600
}

💡Tip: The bearer token is temporary — it expires after the duration shown in expires_in (in seconds).
You’ll need to generate a new access token each time you perform Document Grounding operations in the future (for example, to create, update, or refresh pipelines)

6.c Create a Pipeline

In Bruno, click New Request under your DGcollection.

Method: POST
URL: Your mTLS URL from the Document Grounding Service Binding, followed by /pipeline/api/v1/pipeline
Example: https://mtls.rage.c-1111.kyma.ondemand.com/pipeline/api/v1/pipeline
Click Create.

Add Headers

content-type: application/json
Authorization: Bearer <your_access_token>
Example: Bearer eyJqa3UiOiJ…………….odHRwczovL2F6cjJpe

Add Body

Change the Body type to JSON and paste the following:
The destination is the name of the "Destination" created in your Joule subaccount" (Step 4.c)

{
  "type": "WorkZone",
  "metadata": {
    "destination": "HGDGBWZ924"
  }
}

Click Save, then click → (Send) to execute the request

If the request is successful, the response will display a pipelineId.
This confirms that the Document Grounding Pipeline has been successfully created and linked to your SAP Build Work Zone, advanced edition workspace. Copy the pipelineId from the response (E.g: pipelineId: aa11-bb22-cc33-dd44)

Setup Complete

Your Document Grounding setup is now complete.
All configurations between SAP Build Work Zone, advanced edition, Joule, and Document Grounding have been successfully established.

Beyond the initial setup, you can perform additional operations using Bruno or cURL, such as:

Get all pipelines – retrieve a list of existing Document Grounding pipelines.
Delete a pipeline – remove an inactive or obsolete pipeline.
Trigger a pipeline – manually refresh or re-index workspace content.

These advanced operations will be detailed in the Appendix section in the end.

Step 7: Verify and Test Joule

Now that the setup is complete, it’s time to test Joule. If you already have Unified Joule set up, you can launch Joule directly from the application. (E.g: Successfactors, Ariba, S/4 Private Cloud, IBP etc.)

If this is a new setup, launch Joule by navigating to Subaccounts → Instances & Subscriptions → Joule Instance, click the instance name to open it, and when you see “The service is up and running,” copy the URL and append /joule — for example: https://rigunifiedjoule-test-sge254ddd.eu10.sapdas.cloud.sap/joule

Before launching Joule, create a Role Collection under Subaccount → Security → Role Collections, add the end_user role, assign the role collection to your user in Subaccount → Security → Users.

Ask Joule questions related to your SAP Build Work Zone, advanced edition content (files, wikis, blogs, or knowledge base articles) that were included in the grounding configuration. If everything is working correctly, Joule should return context-aware answers derived from your organization’s grounded documents.

Step 8: Appendix

After completing the setup, you can manage and maintain your Document Grounding pipelines using additional API operations in Bruno or cURL. These include viewing all existing pipelines, deleting or re-creating a pipeline, manually triggering executions, and checking their status.

8.a. Get All Pipelines

Use a GET request to retrieve all existing Document Grounding pipelines in your Joule subaccount. This allows you to verify which pipelines are active and view their associated destinations.

In Bruno: Select your collection (DGcollection) → click New Request → name it Get All Pipelines
Enter the MTLS URL followed by /pipeline/api/v1/pipeline
E.g: https://mtls.rage.c-1111.kyma.ondemand.com/pipeline/api/v1/pipeline

Add headers, save, and click → (Send😞 You will see all existing pipelines
accept: application/json
Authorization: Bearer <access_token>

Note: You need to get a new Bearer token if the previous token is expired. Check step 6.2 to get a new Bearer token

8.b. Delete a Pipeline

Use a DELETE request with the specific pipelineId to remove a pipeline that’s no longer needed or was incorrectly configured.

In Bruno: Select your collection (DGcollection) → click New Request → name it Delete Pipeline.

Enter your mTLS URL followed by /pipeline/api/v1/pipeline/<pipelineId>
Example: https://mtls.rage.c-1111.kyma.ondemand.com/pipeline/api/v1/pipeline/aa11-bb22-cc33-dd44

Add headers, save, and click → (Send):
content-type: application/json
Authorization: Bearer <access_token>
Once executed, the pipeline with the given ID will be deleted successfully

8.c. Trigger a Pipeline

Use a POST request to manually trigger a pipeline execution. This can reindex or update metadata for all grounded documents in the selected pipeline.

In Bruno: Select your collection (DGcollection) → click New Request → name it Trigger Pipeline
Enter your mTLS URL followed by /pipeline/api/v1/pipeline/trigger
Example: https://mtls.rage.c-1111.kyma.ondemand.com/pipeline/api/v1/pipeline/trigger

Add Body

{
  "pipelineId": "aa11-bb22-cc33-dd44",
  "metadataOnly": false
}

Add Headers, save, and click → (Send):
accept: application/json
Authorization: Bearer <access_token>

Once executed, the pipeline will start executing.

8.d. Check Pipeline Execution/Grounding Status of the documents

To check status of individual documents (files, wiki, knowledge article etc.)

In Bruno: Select your collection (DGcollection) → click New Request → name it Check Grounding Status.
URL: https://<your-mtls-url>/pipeline/api/v1/pipeline/<PipelineID>/documents
Example: https://mtls.rage.c-1111.kyma.ondemand.com/pipeline/api/v1/pipeline/aa11-bb22-cc33-dd44/documents

Add Headers, save, and click → (Send):
accept: application/json
Authorization: Bearer <access_token>

8.e. Check Pipeline Status & Total executions

Replicate steps from 8.d and use "status" and "executions" instead of "documents" in the URL. You will get final status or total number of executions of the pipeline

URL: https://<your-mtls-url>/pipeline/api/v1/pipeline/<PipelineID>/status OR executions

Example: https://mtls.rage.c-1111.kyma.ondemand.com/pipeline/api/v1/pipeline/aa11-bb22-cc33-dd44/status
Example: https://mtls.rage.c-1111.kyma.ondemand.com/pipeline/api/v1/pipeline/aa11-bb22-cc33-dd44/executions

SAP Document AI accelerates document-centric business processes with Generative AI

2025-11-10T13:54:41.812000+01:00

Authors: @amagnani37 & @merza

Thank you to everyone who joined our recent partner webinar, "Streamline document processing with SAP Document AI". This session was part of our ongoing partner webinar series, "Talk to your business with Generative AI", and the turnout and engagement were fantastic.

For those who couldn't make it, or for anyone who wants a recap, we—Merza Klaghstan, Alice Magnani, and Jacob Tan from the BTP & AI Solution Architects team —wanted to share the key takeaways, demos, and technical insights from the session.

Our goal was to go beyond a simple feature list and show you how SAP Document AI can be the centerpiece of a complete, end-to-end business solution.

Here’s what we covered.

Part 1: What is SAP Document AI and Why Does it Matter?

We kicked off by tackling the core business problem: a staggering 70-80% of all enterprise information is unstructured. Manually processing this data from documents, emails, and scans is not only slow but also expensive and prone to errors.

SAP Document AI, formerly known as Document Information Extraction (DOX), is SAP's AI-powered solution on the Business Technology Platform (BTP) designed to solve this. It goes far beyond simple Optical Character Recognition (OCR) to extract, classify, and validate data with high precision.

Key Capabilities

We highlighted several key features:

Multiple Editions: The service is available in Base AI, Premium AI (which includes Generative AI capabilities), and Embedded editions to fit different use cases. We also want to highlight for our SAP Partners that the Premium AI edition is part of the Test, Demo, and Development (TDD) license, which provides a much more convenient model for your non-productive and internal development purposes.
The New "Workspace" UI: We compared the classic "Basic" UI with the new, full-featured "Workspace," which is the future of the service and includes capabilities like email channel configuration and workflow management.

It's the new, full-featured environment for all document processing tasks . We want to emphasize that from now on, all the latest SAP Document AI features will be available in the Workspace UI only, so we strongly advise using it for all new projects.
Pre-Configured & Custom Content (Schemas): The service provides many pre-trained models (called "Schemas") for standard documents like invoices and purchase orders. The real power, however, comes from creating custom schemas for any document type your business needs—from specialized licenses to internal forms.
Generative AI & Instant Learning: For custom schemas, SAP Document AI leverages Generative AI and Large Language Models (LLMs) to understand and extract information. We also demonstrated "Instant Learning," where the model learns from your manual corrections immediately. After one correction, it knows how to handle that document layout correctly the next time.

In our demo, we showed how the model initially struggled with a name containing a middle name and extracted the wrong federation name (full name vs. acronym) . After one manual correction and confirmation, it learned immediately. Technically, this feature uses few-shots prompting to make the underlying LLM aware of your previously reviewed documents to improve accuracy on the next run.

Part 2: A Real-World Use Case: "Vertigo Travels"

To make this real, we introduced "Vertigo Travels," a fictional travel agency specializing in sports and adventure holidays. Their key challenge is managing the various kind of documents required for subscriptions, such as ID cards, medical certificates, and diving licenses.

The solution architecture includes:

The Vertigo Travels website, a frontend application, deployed in Cloud Foundry, serving as entry point for Travelers to select travel packages and upload the necessary documents, such as ID cards or medical certificates. The backend logic is a CAP (Node.js) application also deployed in a cloud Foundry.
SAP Document AI, for all document intelligence operations, communicating via API with our backend application.
SAP HANA Cloud, as a persistence layer for our application, to store all the entities involved in the subscription process.
SAP Build Work Zone standard edition, as the entry point for Vertigo Travels employees to supervise the process. From Work Zone, business users can access the Vertigo Travel website for admin operations, or the SAP Document AI workspace, for instance to review documents. SAP Build Work Zone also allows business users to use SAP Document AI via mobile.
SAP Build Process Automation, to define business rules to automate some document validation checks. Although this could also be implemented as part of the Vertigo Travels backend application, SAP Build Process Automation enables citizen developers to modify easily the business logics when required.
SAP S/4HANA Cloud, as the core SAP Cloud solution. The Vertigo Travels backend is integrated with modules such as Business Partners, Products, Sales Orders/Invoice, Customer Return for refund.

For the source code, you may refer here for reference.

P.S. if you're planning to work on this prototype, please give us a shout out~! So at least we know somebody is trying our repository out!

We demonstrated three end-to-end scenarios:

The E2E Flow: A traveler, Mary, subscribes on the website and uploads her documents. A back-office employee, Barry, reviews and corrects one document in the Document AI Workspace. Once confirmed, a Sales Order is automatically created in S/4HANA Cloud.
Multi-Channel Ingestion: A traveler, Luca, visits an office. The employee uses SAP Mobile Start to scan his ID and medical certificate. The system then automatically emails Luca about a missing document. He simply replies to the email with his license attached, which is automatically ingested via the Inbound Channel feature.
Automation: A traveler, Claire, uploads an income statement to apply for a discount. SAP Build Process Automation runs a rule to check her eligibility. When she submits her final documents, they have a high confidence score and are auto-confirmed without any manual review.

Part 3: How We Built It (The Technical Deep Dive)

For our developer community, we walked through the "how-to" for the key components.

For the source code, you may refer here for reference.

P.S. if you're planning to work on this prototype, please give us a shout out~! So at least we know somebody is trying our repository out!

1. Setup and Schema Management: We started with the BTP setup, which involves managing entitlements, creating the service instance, and, critically, establishing trust with SAP Cloud Identity Service (IAS) for user authorization. We then showed how to create the custom schemas that power the application.

When you set up your entitlements, make sure to use the application subscription ending in -ias (Identity Authentication Service), as this is the one that provisions the new Workspace interface.

2. API-Driven Development: Our CAP application's backend communicates directly with Document AI's APIs. We showed code snippets for uploading a file (using the FileService) and processing it against a schema (using the SchemaService).

How to retrieve specific extracted values:
While our demo UI showed a simple text extraction, your application will likely need specific, structured data (like "gross_family_income" or "number_of_kids"). The API makes this straightforward:

After processing the document, you get a Document ID.
You can use this Document ID to retrieve the Document Version ID.
Using the Version ID, you can then query the Entities for that document. 4. Finally, you can retrieve the specific value by using the Entity ID (e.g., the ID for your "number_of_kids" field) .

This is how our automation process, for example, was able to pull only the family income and number of kids to pass to the rules engine.

3. Configuring Email Ingestion: We configured the "Channels" feature to monitor an email inbox. This required registering an app in Microsoft Azure, granting Mail.ReadWrite permissions , and then using the Document AI Workspace to map a specific email folder to our target schema.

4. Enabling SAP Mobile Start: This was a key integration for the in-office scenario. It involved three steps:

Adding the Document AI dependency to SAP Build Work Zone within the IAS console.
Creating a specific BTP destination (sapdocumentaimobilestartintegration).
Configuring the app tile in the Work Zone Site Manager with the correct navigation and visualization parameters (e.g., mobilestart.type = documentAi).

4. Automating Logic with SAP Build Process Automation: To keep our application logic flexible, we externalized our validation rules. We created a process in SAP Build Process Automation with a Decision Table (Rule) to calculate the family discount based on income and number of kids. This process is exposed as an API and triggered by our CAP application.

For the source code, you may refer here for reference.

P.S. if you're planning to work on this prototype, please give us a shout out~! So at least we know somebody is trying our repository out!

Part 4: Conclusion, Best Practices, and What's Next

We concluded with some key advice and a look ahead.

Best Practices & Responsible AI

Quality In, Quality Out: The accuracy of extraction heavily depends on the quality of your scans.
Human in the Loop: While automation is powerful, always design your processes with a "human in the loop" for validation, especially for critical data.
Responsible AI: We reiterated SAP's commitment to AI ethics. External LLM providers are carefully selected and are not permitted to store or retrain their models on customer data. Features like continuous data sharing for model improvement are strictly opt-in and should not be used for sensitive personal data.

Get Started and Look Ahead

We are committed to helping our partners build on this platform. Here are the resources to get you started:

SAP Document AI Official Help Page: https://url.sap/imohkv
Vertigo Travels GitHub Repository: https://url.sap/pi5vwn (The code for our demo app will be available here!)
SAP Community for AI/ML: https://url.sap/ai-ml-grp

The roadmap for SAP Document AI is packed with features, including more pre-built schemas, document workflows, custom LLM prompting, and deeper integration with Joule.

Finally, don't miss SAP TechEd Virtual! Be sure to check out sessions AI821v and AI102v for the latest roadmap and deep-dive innovations from the product team.

Please feel free to leave any questions or feedback in the comments below!

SAP Document AI Made Simple: A Functional Analyst’s Plug-and-Play Guide

2025-11-13T22:15:06.916000+01:00

My initial plan was to se Boosters for Free Tier Use of SAP AI Core and SAP AI Launchpad
(Be aware- AI core and AI launchpad are not available in BTP free tier. Alas!)

Next Goal:

Build an end-to-end document AI demo that:

Uploads an invoice or receipt (PDF/image)
Extracts structured data using SAP Document AI API
Stores the response as JSON
Visualizes extracted fields in Streamlit

Architecture

[Streamlit UI] → [BTP Destination/Service Key] → [SAP Document AI API]

↓

[Response JSON → Results Display]

But hang on – If you are a first time user of SAP BTP, ( Most of SAP business analyst, Functional consultants, SAP managers in 2025 are right at this phase- they all have heard about BTP, knows what it is used for – but never used it hands-on ) there are too many pre requisite one after another. More interesting fact- not all of them are in a single place. You need to stitch bits and pieces to make it work.

I researched few articles over internet even asked Walkme- but didn't get necessary answers

Assumption - Most of the folks who would read this article have SAP S-user ID from past interaction with SAP marketplace portal.

Note – Reference: https://developers.sap.com/tutorials/hcp-create-trial-account.html

To start with- Access SAP BTP Cockpit and create Global + Sub account under free BTP trial.

Once the signup process is complete, you can access the cockpit via https://cockpit.hanatrial.ondemand.com/trial/

Choose region- Either of the below- it will take up to 3 minutes to create BTP accounts.

This page below will display the current state of the subaccount. You can manage your subscriptions and jump into the different runtime environments. It also shows you fundamental; information of the Cloud Foundry environment, such as the API endpoint and the available spaces – SAVE it as favourite (Top right corner star icon inside SAP BTP)

You may need to navigate between Global and sub account and different nodes often.

A SAP BTP subaccount is best described as a workspace for development and application deployment that provides strong isolation for security, user management, and data segregation, rather than a "client" in the traditional SAP multi-client sense.

We can have multiple sub account inside one global BTP account

Next step- Run Booster method to enable SAP document service

NOTE - We do not need to Manage Entitlements Using the BTP Cockpit.
Rather we shall use booster method under Global BTP account
Go to global BTP account page- click on left panel boosters as shown below

scroll down – under artificial intelligence category

This booster helps you set up SAP Document AI in the SAP BTP cockpit.
Step by step, this booster assigns entitlements, updates your subaccount, and creates a service instance and the associated service key for SAP Document AI.

Click start

Reminder- If you close SAP BTP browser by mistake – here is a quick link to open it again - https://cockpit.hanatrial.ondemand.com/trial/

Next go to BTP sub account – which is visible under Global account as below- left panel you can click on account explorer

Inside sub account ->expand services - look for instances and subscriptions on left panel- shown below

We can see 2 subscriptions below by default SAP Business application studio is enabled

Next click on service marketplace

Search with AI + Enter- select 1st tile shown below

Click on Create (as I pre-created so it is greyed out)

Fill up below popup

Tips: You should select the plan as shown below to use the SAP AI document service effectively as a plug and play from external ecosystem.

Setup your new instance as below- name as per your choice

Next – From left panel, again go to instances and subscriptions- this time 2nd tab called instances

Note – the difference in “Plan” in the setup, one is default another is blocks of 100.

Click on the credential “1key “ and 'Form'- we explain this in details. You may download it.
1/ Url -i.e. DOC_AI_URL used by external application.

2/ TOKEN_URL = This will be needed by external apps to authenticate with BTP service.

This url has term ‘authentication’- -Tips- do not use sburl

Tips- when using this Token for authentication, we need to add suffix befind above url as below after ondemand.com
/oauth/token"
3/ CLIENT_ID – That is self-explanatory- from the above screenshot, you must note down.
4/ CLIENT_SECRET - That is self-explanatory- from the above screenshot, you must note down.
Inside the key file there is another element called swagger

When we would make a call to SAP AI document services- our schema from external applications, then this swagger needs to be passed along with the document Url.

Full API endpoint (for extraction jobs)
combine "url" + "swagger"

Next- From left panel -click on Security- still within BTP sub account

Look into right panel – click on arrow – to navigate inside details

Click on the Role collections tab- It will show already assigned existing roles- do not get confused or search roles here. Click on assign role collections

In the below popup-search for ‘doc’ and assign all of them to your BTP user

Note- without those roles you will not be able to use SAP AI doc service schema configuration- neither the external call will work.

Tips- this is crucial step-otherwise SAP BTP service call will authenticate but not be able to use your own custom schema or SAP defined standard schemas and purpose of the plug & play will be defeated.

Interestingly this role collection step you may have to do multiple times as I observed even after assignment on 1st day- next day it vanished- had to re assign.

Next go back to Instances & subsc.from left panel

Click on application = SAP doc AI shown below

A new browser widow will open as below

Tips- Without suitable role collection this arrow marked button will not appear- This is used for schema creation.

Click on schema configuration Below list of schemas will be shown- SAP_ are all standard schema provided out of box-we can use it as is or create our own schema. When we click on create – below popup appears

As shown below I created my own schema namely invoice_ai_doc

I clicked on my schema name- copy of SAP_invoice_schema Next – I click on create version- I could now get ‘Edit’ button enabled for both header and item fields of the invoice schema- I can click on ADD button to add my bespoke fields as shown below.

Finally click on activate the schema
Now all necessary BTP setup is ready.
Next- outside SAP ecosystem – plug and play.
For simplicity, I shall use ‘Streamlit’ an open-source Python library that allows us to create interactive web applications using pure Python and a python project on PyCharm IDE
Note- Any web application/tech. stack React, Django, Flask, JavaScript Node.js backend, Angular, CSS etc. can leverage SAP BTP Document AI service with above setup.

I created following python project in PyCharm - where I loaded secret key of SAP BTP DOC AI service instance in a safe and secure way. I have used .env and .gitignore to protect unwanted github publishing of the key details as per best practice guidelines.

A simple object class is created for this demonstration

Having 4 methods as below

Next showtime! I run my streamlit app UI called app.py(see python project)

The frontend is displayed below

Feature-
1/ I can test service configuration with SAP BTP authentication token
2/ Available schema
3/ Browse & Upload invoice and click ‘Extract document information’ on streamlit app
a) I upload a train ticket invoice PDF

b) A sample invoice jpeg
c) A sample invoice pdf with different details.

Result –

2nd Invoice image

3rd Invoice also uploaded for analysis and extracted on my app as below

Next – check on SAP BTP portal, where we submitted those 3 invoices (2 pdf and 1 jpeg)

Below is the document service page, amazing!

I can also open any invoice – example below – It also shows ECR values right side!

Conclusion:

In a world where technology evolves faster than documentation can keep up, it’s easy for Functional Analysts to feel lost and overwhelmed. This guide bridges that gap, providing clear, step-by-step instructions to integrate SAP Document AI seamlessly with any non-SAP web application. By removing the guesswork and eliminating outdated instructions, it empowers analysts to explore, experiment, and innovate without frustration. The result? Less confusion, more confidence, and the ability to unlock the true potential of intelligent document processing in your organization.

How to build an AI side by side extension for Fieldglass

2025-11-26T15:00:00.024000+01:00

In this blog post, we’ll walk you through building a side-by-side extension that combines SAP BTP’s AI capabilities with Fieldglass to streamline external worker onboarding. Using a practical example, we demonstrate how to create an intelligent document validation system that automatically reviews content of worker documents (ID, criminal records, debt collection statements, and confidentiality agreements) against company guidelines. The AI-powered extension extracts key content from documents, validates it using large language models, and presents summarized findings to end users for final approval or rejection decisions. This approach with GenAI Hub to protect these sensitive information significantly reduces manual review time while ensuring compliance, with all decisions automatically documented back in Fieldglass for complete audit trails.

In this example we test with for types of documents:

ID
Debt collection
Criminal record
Confidential agreement

Architecture:

Document AI: Service to retrieve the information from the different types of documents. For each document there is a specific schema defined

HANA Cloud: Stores all the needed data for the workflow/UI. Possibility to compare document to reference document. Also stores the guidelines that are used to check the entries from the documents

SAP AI Core: Access to the different LLMs. Used to decide if the entries found with Document AI are acceptable per guideline or not. Summarizes the findings for the end user and gives recommendation

Destination Service: Stores the credentials for OAuth2 and other relevant API data to access SAP Fieldglass from the CAP application. You need to generate an API key for Fieldglass (sap help documentation)

Process Flow:

Here is a very basic overview of the process flow of the application:

We retrieve all workorders from Fieldglass and for all new ones we also retrieve the documents. If all documents are available for a work order we start the validation process. Validated will be ID (Name, birthday), confidential agreement (signed not signed), debt collection and credit statement are checked for entries. Results are validated over AI Core based on company guidelines. Resulting summary of findings are being displayed in the Fiori App to decide. The decision is then being sent over API to Fieldglass with a approval comment or a reject reason. The decision is being documented in Fieldglass. For each document type there is an Activity Item which has the verification setting enabled so we can accept, reject it

User interface:

The user can see the summary of the users entries, the current item being validated and the items that already have been validated. The user can also look up the original document to cross check the summary if needed.

APIs and endpoints:

Fieldglass APIs:

The Fieldglass APIs are, to be honest, not that straight forward 😉

For this example we used the following APIs:

Activity Item Completion Upload (help, API docu)

To set the status of the Fieldglass Activity Items as Rejected, Completed or Verify and with the reject reason/comment

Work Order Delta Download (help, API docu)

For retrieving all the work orders. Be aware this gets you a delta since the last call! Not all Work orders

Worker from Work Order Download (help, API doc)

Fetch all information of the worker (name, security level etc.)

Attachment XML Download (help, <no API available in business accelerator hub)

For retrieving all relevant documents from a specific work order
Parameters need to be set like the following

?object_ref=Activity Item~${workOrderID}~${activityItem}
The "activityItem" variable is a code you need to look up in the admin configuration and specifies what Activity Item you want to retrieve: Go to Admin Configuration -> Activity Item -> then search for you Activity Item and get the according code

Be aware of the Fieldglass APIs ;D

A lot of the FG APIs accept only csv in the request body. Make sure you sent the correct value in the correct cell. Use for example Excel if you want to make sure that you entered the correct values underneath the correct header as csv can be hard to read in just plain text and you can easily enter your value in the wrong column
They can have spaces in the request URL. Make sure they are correctly encoded by the REST client used
Some APIs do not make a full read but only a delta of the underlying entities

You can reset the last fetch in Fieldglass if needed for testing

In General be very aware of the API syntax as it can be unusual when comparing to the structure of other APIs. For example the syntax of the parameters for the Attachment XML Download with the "~" in between. E.g. for the download the parameter differs greatly depending if you want attachments from an Activity Item or a SOW

In conclusion read the documentation very carefully on the business accelerator hub AND the help portal

How to call the Document AI Service:

To get the URL of the we bind the Document AI Service to our application and used the function "getServiceBinding" to retrieve all the required properties (@sap-cloud-sdk/connectivity package)

For starting a job in Document AI you need to set the following headers and make a POST Request:

Header field	Value	Desciption
clientId	<clientId>	You can call this Api to retrieve all client ids
documentType	<documentType>	The type of document being uploaded. Choices: 'invoice', 'paymentAdvice', 'purchaseOrder', 'businessCard', 'deliveryNote', 'resume', 'birthCertificate' 'custom' if you created one yourself
schemaId	<schemaId>	The schema you want to use for your document. (get all schemas with this api)
schemaVersion	<schemaVersion>	The version of your schema
Content-Type	Multipart/form-data

As a response you should receive the jobId this jobId is needed to call the service again later for the result.

In our case we configured the callback from the service, so it calls our application when the job is finished and all results have been extracted. Check this out, to see how this is done. Make sure you have created a configuration key and set "activateDocumentNotifications" to true otherwise the service will not perform a callback. Configure the destination for the callback accordingly.

When the service performs the callback with the job id you can use this to make a GET request to fetch the results of the job. This information is then used to further process the documents and later sent to AI Core.

You can use predefined schemas or you can define your own schemasd. In our case we defined custom schemas. Check out the help portal for schemas

How to call the AI Core

For using the AI Core you can use the provided SDK which fits you best. In our case as we are working with CAP and a NodeJS backend we used the SAP Cloud SDK for AI (JavaScript). We then utilized the Orchestration Service which enables us to quickly change back and forth between the difference models if needed

Set up for side by side extension with Fieldglass:

All APIs of Fieldglass are available in the business accelerator hub and in the help portal

For making HTTP requests to FG you need to have an API Key. Here is a short Joule for consultant explanation on how to do that or check out this help documentation:

Step-by-Step Instructions for API Key Generation:

Access Configuration Manager Dashboard
1. Sign into SAP Fieldglass using your standard credentials$
2. Switch to Configuration Manager user: Click your user icon in the upper right corner and select "Linked Accounts"
3. Navigate to the Configuration Manager dashboard - this provides access to integration tools and administrative functions
Create the API Application Key
1. Click the "API Application Keys" tile on the Configuration Manager dashboard
2. Click "New" to create a new API key
3. Fill out the form with the following required information:
  1. Application Name: Enter a unique, descriptive name (e.g., "SAP Start Integration")
  2. Description: Provide a clear description of the API key's purpose (e.g., "API Key for SAP Start Integration")
4. Click "Create" to generate the API key
5. Capture Generated Credentials

After creation, the system will display auto-generated values:

Application Key (auto-generated)
Client ID (auto-generated)
Client Secret (auto-generated)

!!! Important: Copy these values to a secure location immediately, as you cannot retrieve them later !!!

BTP Destination:

Create a BTP destination with the url, credentials and Oauth Url from the API key. Choose "OAuth2ClientCredentials" as Authentication. Don't know how to create a destination? Check this out!

You can now call the SAP Fieldglass APIs over this destination. In case of errors try to call SAP Fieldglass with an REST Client like Postman or Bruno

You can retrieve the destination in CAP for example with the @sap-cloud-sdk/connectivity package with the getDestination function and then make an axios call to Fieldglass

Conclusion:

Congratulations! Using the example of an AI-powered external worker document validation system, you’ve learned how to build a side-by-side extension for Fieldglass with SAP BTP’s AI services. Through this practical use case, you now understand how to leverage SAP Document AI for intelligent document processing, SAP AI Core with its LLMs for automated guideline validation, and navigate the unique challenges of Fieldglass APIs - from CSV request bodies to URL encoding quirks. With this real-world example as your foundation, you’re ready to create your own intelligent automation solutions that enhance Fieldglass 🚀

Special Thanks:
Big thanks to Leo Irudayam for his dedication, expertise and hands-on development that helped bring this PoC to life.
Special thanks also goes to Matthias Beeler for the technical implementation and realization and Sven Breuninger and Kevin Hornung for the Fieldglass expertise that made this PoC possible.

RPT-1, Document AI, Perplexity in AI Core and co. - How to start your 2026 right with SAP's AI offer

2026-01-05T13:50:32.357000+01:00

AI remains years after its new wave of generative AI more omnipresent then ever in our daily lives and there not a single day passes where we don't talk about new announcements, AI ethics discussion, if it is a bubble, what amount of money is spent and so on. For those of you, who follow me, will know, that I preach all day that the key to successful AI adoption is a useful case, the right technology and working with the people.

Writing prompt guidelines is just enablement for another side-by-side tool and does not gain efficiency rather joy trying out new things. Building an agent for a pretty deterministic process might be fun and easy to implement, but expensive and unpredictable in production. And so on... While many are talking about large mainly American software vendors with their "agentic foundation killing any SaaS", SAP's offerings might not be so familiar to you (besides Joule and BDC) - so that's my motivation to provide a sneak peak of exciting topics in 2026.

Use Case: Onboarding of a new vehicle to the car fleet

In order to make tools tangible and understand the bigger picture, we will sketch an AI-assisted process of vehicle onboarding. Previously following steps were mandatory:

Clerk typing vehicle data into system from the official vehicle document
Vehicle Master Data Engineer adding additional information based on public search as required drivers licence and consumption values
Finance expert to determine the tax amount to be paid for that year based on manual and local review
Field Service Manager classifying the vehicle for the right purposes to be used

As you can see, this process is slow, very labor-intense and quite possibly error-prone. We will now step-by-step automate this by leveraging different AI tools offered by SAP within the Business AI portfolio 🙂

Our ultimate goal is creating a low-code-first, enterprise-ready automation to avoid manual research and typing into the system:

Capturing of correct data: No more manual data entries from 2026 on

If you have in 2026 tasks which require typing in data from one medium (e.g. mail or paper) into your system, it's time to automate this asap. In our scenario, we do have the vehicle document on paper, as this one:

Instead of typing this, we use the SAP Start app on our iPhone to take a picture and the deep integration to SAP Document AI to process this document (see more here). We then utilise Document AI with our custom schema to extract the results and preview these to our clerk for a double-check. Instead of building entire new UIs, workflows, etc. we can now do this directly in Document AI and get a nice view of the document on the left and extraction on the right:

Automating web searching: Leverage the perplexity integration in SAP AI Core

SAP AI Core is the central hub for any access to MLOps activities and GenAI. With your BTP account you have access to a wide variety of partner models hosted on hyperscaler infrastructure and SAP's very own one with Mistral. If you didn't know this yet, that we have an orchestration workflow with prompt optimiser and much more, look up here which models we offer through your BTP contract.

Lately, we have added Perplexity as well - the AI-assisted search engine. Now we can connect publicly available data straight into our workflows, and it's huge, e.g.

classify the purchased non-catalog item to the correct product group
fill gaps in master data based on product sheets
make research on sustainability KPIs of your supplier
conduct analysis for B2G processes
get market macro trends or real-time insights, e.g. supply chain disruptions
...

We will leverage the Perplexity integration to enhance our vehicle master data by the required drivers licence type for Swiss drivers licences and the energy consumption according to WLTP (an industry standard to measure consumption). We can model our prompt in SAP AI Launchpad and use the prompt registry to save it with placeholders:

And now it's so simple to add it to your pro-code app via CAP, to your low-code automation in SAP Build Process Automation or your integration flow in SAP Integration Suite:

In this iFlow you can see how we leverage the new outbound channel feature in Document AI to enrich the extraction of the results straightaway with an iFlow exposed via API management (for more, stay tuned on upcoming webinars). If you have seen in the previous screenshot of Document AI - there is a tab with "Z - additional values" 😉

Do machine learning without data scientist: SAP's own foundational model RPT-1

Now we have been onboarding our vehicle, but we must classify for which tasks we can use it. Luckily we have already quite a large fleet and some data. Ideal for machine learning (yes that's also part of AI and yes AI is not only GenAI).

Machine learning learns from historical data - as simple as this is, this requires a data set, model fitting, model validation and deployment. Not straightforward. With Gemini, Le Chat and co. we are used to just prompt AI models. What if, we could just put a snapshot of our table to a pre-trained model and let it classify by itself. Congrats - you are now a RPT-1 user:

RPT-1 has its own playground, is accessible via AI Core and is phenomenal for transactional classification and prediction tasks. If you want to know more about its integration, I'd recommend you this gem. In my scenario, we use RPT-1 in our workflow in Build Process Automation to classify our vehicle correctly before we post it to our equipments in S/4HANA:

Performing more agentic tasks: Joule Studio

Previously you might have heard, you can use Joule Studio to make Joule truly yours. Joule Skills are intent-based automations made familiar to Joule to perform tasks conversational based. They can leverage workflows, pro-code and integrations via Integration Suite.

2025 Christmas gift, was the GA of the agent building capability in Joule where we can model agents. Agents are behaving non-deterministic and trying to mimic a human thought by providing it a scene and background - like an improvisation theatre. Now it's oftentimes pretty tricky to understand when an agent is really necessary. In our example we use an agent to determine how the vehicle tax are calculated for the Kanton St.Gallen, where we have registered our car and perform a calculation based on the perplexity web search (disclaimer - this can be modelled through a workflow in Build, it's just for the sake of simplicity):

Why choose SAP for your AI journey?

That's a good question and especially given there are soooo many tools on the market, why should you? It's simple: it's all about enterprise readiness. Behind every of this automation, there is only 5-10% AI, the remaining parts are security, audibility, traceability, transport management, scale resiliently in production, monitoring, low-code modelling etc. And for all of the remaining bits and pieces, SAP BTP has always been the best option from SAP Build to SAP Integration Suite. With SAP Document AI, we have a leader in document processing with inbound and outbound channels, and SAP AI Core a powerful AI orchestration service at low costs each.

While you can get other services cheaper somewhere, the TCO of this toolbox is imho astonishingly low. And for building this scenario from mobile device intake to equipment in S/4, I just needed a day to configure and build without pro-code needed. Happy AI-2026...

From Docs to Dialogue: Unlock AI Self-Service Learning for Your Internal Teams

2026-01-08T23:27:21.357000+01:00

From Docs to Dialogue: Unlock AI Self-Service Learning for Your Internal Teams

Every organization wants onboarding to be faster, clearer, and less dependent on tribal knowledge.
NotebookLM enables the conversion of trusted product documentation and internal project artifacts into an AI-based conversational resource that employees can engage with naturally, as if they were chatting with their most knowledgeable coworker. Instead of searching across files, PDFs, and web pages, teams can ask direct questions about product behavior, setup requirements, and configuration considerations, and instantly turn answers into internal training materials.
This AI-driven approach
- scales knowledge without replacing expertise
- streamlines learning without adding process overhead
- helps organizations build deeper product fluency internally

Table of Contents

Benefits & Use Cases

Faster employee ramp-up by turning existing SAP Documentation into a conversational AI learning environment
Instant answers for distributed teams across roles, experience levels, and time zones
Improved configuration confidence through real-time dialogue with authoritative sources
Consistent onboarding experiences that reduce knowledge gaps between new hires
Rapid training asset creation (FAQs, slide outlines, learning summaries) generated directly from source material
Search less, learn more—teams ask questions instead of digging across files and folders.
More informed internal experts because foundational questions get answered efficiently and consistently
Support interactions become more efficient and solution-focused when they happen, backed by deeper internal product fluency.

Gathering Trusted Knowledge

To begin using NoteBookLM as your AI Self-Service learning tool, you must first gather all of the relevant knowledge on the specific SAP product you are looking to focus on. These will become your sources within NotebookLM and will act as the knowledge base that NotebookLM uses as it guides your

The free version of NotebookLM is limited to 50 sources, so please consider this as you capture your list of sources

SAP Product Documentation

https://help.sap.com/docs/ contains all relevant product documentation for your SAP products and is the best place to start gathering the content you need.

Go to SAP Help
Search for the specific SAP product you are looking to hone in on and then click on the Suggested Product that most aligns with your search.
Example: SAP Business Network for Procurement
In the bottom right corner of the top banner, click Download PDF
Select all relevant options from the pop-up page, then click Download at the bottom of the page

Getting Started with NotebookLM

Now that you have all your sources prepared, you can begin uploading them to NotebookLM.

Open NotebookLM and complete the onboarding process to create a free account. This will be tied to your Gmail/Google account. Please consult your internal teams regarding any enterprise plans or privacy concerns.
Create a new notebook, selecting the + Create New button
Upload the files that you captured in the previous step by clicking Upload Files
Allow your sources to load. Once complete, you should see all your sources displayed in the left panel.
To share your notebook with other colleagues, you can click the Share button in the top right of the page.

Chatbot-Driven Learning

In the middle section of the page, you can interact with the notebook via Chat, as if you were talking with a product expert.

NotebookLM will then use AI to develop an answer, intentionally prioritizing your repository of sources to provide the proper context and deliver the best answer to your question.
Within the AI-generated response, you will notice that small numbers with circles around them are displayed throughout the answer. You can click these, and then the left panel will take you directly into the specific source document where it pulled that part of the answer from
For a more visual approach, you can also click the Mind Map option in the Studio panel. Once generated, you can click to view a visual representation of the sources you uploaded. As you click deeper into the mind map, the chat will auto-update and help kick-start a conversation you can use to drill down into a specific topic area.

As with any AI LLM-based tool, results may vary and should be reviewed in detail by your teams for accuracy.

You can continue your conversation with the AI chatbot by asking it

additional clarifying questions specific to your needs
reframing your question in more complete wording to provide more context
to drill down deeper into specific areas of its response
to drill down deeper into specific documents/sources
Additional documentation

Generating Internal Training Assets

The right portion of the page, called Studio, will include options (depending on your plan) to generate content tailored to your team's training and enablement needs.

For each option in the studio, click the Pencil icon to customize the result by providing a prompt that guides the studio's content generation and is tailored to your needs.

If you do not click the pencil, it will auto-generate based on all your sources. Depending on the size and scope of your sources, the output may be too high-level without a custom prompt provided.

Each Studio option may also provide its own set of configuration toggles in addition to a custom prompt.

Each option also includes a Download option, allowing you to export any generated content to your internal libraries or training tools.

Audio Overview: podcast interview for auditory learning

The audio overview option generates a one-on-one podcast interview between two people who discuss your sources and/or your custom prompt. You can use this to create targeted podcasts that dive into specific areas, then include them in onboarding materials for teams who prefer an Auditory format.

*Pro Tip: upload your own transcript or outline of how you would like the conversation to go. This will help guide the model to create an interview that covers the areas you consider essential, rather than guessing what to highlight.

*Pro Tip: When providing your custom prompt, include specifics as to how you would like the tone of the speakers to be. By default, the podcasts may seem a bit exaggerated at times. You can ask the prompt to use more normal and realistic tones and/or ask it to avoid certain words like "massive" or "huge", for example

Infographic: a one-stop shop PDF for visual learning

The infographic will create a one-pager PDF with AI-generated visuals that provide a more "connect the dots" overview of the sources. It will start with a high-level overview that connects all sources (similar to a Mind Map), but if you provide a specific prompt, it can also provide a more refined overview of that topic area.

*Pro Tip: check the visual for any discrepancies. If you see issues, recreate the infographic by changing your custom prompt to call out the mistake you want to avoid. The more specifics you provide, the better

*Pro Tip: In your custom prompt, call out the audience that will be consuming the graphic itself. It can tailor the messaging and graphic style to better suit that audience. You can also ask it to reference a branding source if you have uploaded one that includes logos, color schemes, etc.

Video Overview: a training video for visual learning

The video overview will provide a more traditional training format, with one speaker presenting the topic while an AI-generated video plays in the background. The video will resemble a PowerPoint presentation. You can use this to create a high-level video overview of the sources, or to provide a custom prompt that goes into one area in more detail.

*Pro Tip: The graphics or numbers on the video itself need to be reviewed for accuracy. If you see discrepancies between what is displayed and what is mentioned in the video, you will need to refine your custom prompt and regenerate the video to close the gap. The more specific you are (e.g., an outline or script and dos and don'ts), the better the results you will get. You are unable to edit the video itself or adjust the slides/visuals after it is generated

Flashcards and Quizzes: A hands-on tool for your team to learn

The Flashcards option searches for primary and recurring keywords and definitions in your sources and automatically generates flashcards to help learners study/memorize key terms.

The Quiz option will also search for primary and recurring keywords across the sources and generate an interactive question-and-answer quiz.

*Quizzes cannot be downloaded externally

Best Practices

Depending on your sources and the sheer volume of data within the notebook, the AI Models can and will hallucinate. Having detailed, specific prompts to guide each conversation or studio creation toward a specific goal will produce the best results. It is better to break your content needs into smaller groupings with detailed prompts than to blindly ask it to create high-level overviews.
Be sure to consult with your internal privacy and compliance teams before uploading any non-public information into NotebookLM. As a general rule, only upload publicly available content to NotebookLM.
YouTube videos can be effective at providing an additional perspective through a more expert-focused, human-guided lens on your dataset of sources. Collect long-form YouTube interviews with experts in your area and upload the URLs as additional sources to your notebook.
Create your own scripts or outlines beforehand and provide those to the chat or studio in addition to your primary ask it. This will help ensure it follows the correct path you are looking for and does not hallucinate. You can also use other LLMs or AI Chatbots to help you come up with ideas and draft a high-level script or outline to follow.
If you share your notebook with end users, ensure they understand that chatbot results must be reviewed and refined to provide guidance that helps kick-start an investigation into a topic. End users should be wary of copying/pasting AI-generated content externally.

End-to-End overview

Find your product at SAP Help

Download the SAP product documentation to PDF

Create a notebook and upload files to NotebookLM

Observe your sources and begin interacting with the chatbot & studio

A real-world example, see NotebookLM in action

The SAP Business Network for Procurement is used globally to connect buyers and suppliers and enable interaction and transactions. This results in a variety of country-specific invoicing laws and portals that vary by country. The application itself includes standard invoicing features and built-in support for various country-specific E-Invoicing mandates.

In this example, we will investigate how we can learn more about this topic using NotebookLM with a single prompt to generate a text answer, an infographic, and a video overview file.

Prompt

I want to learn more about e-invoicing and country-specific invoicing rules within the SBN for the Procurement application. My team has a basic understanding of the core SBN Procurement flows, but is seeking more guidance on the e-invoicing and country-specific invoicing logic and features within the product. Can you help produce content that clarifies this for us and explains the topic in detail as it relates to the larger product? Id like to know what is supported in the app, how i would interact with these features, if there are any specific setup or customization requirements to use this, and a list of which countries are currently supported. 
The audience here is a group of corporate employees who know procurement as a whole. Keep things professional and short/sweet. Don't use any exaggerations and don't overhype with any big, bold words. Use simpler, down-to-earth professional wording.

Infographic

Video Overview

Chatbot Response

Here is an overview of how SAP Business Network (SBN) handles e-invoicing and country-specific compliance.

In the context of the Procurement application, e-invoicing on SBN goes beyond emailing a PDF. The system uses structured data (cXML) to validate invoices against your business rules and the legal regulations of the supplier’s country before the invoice reaches your internal systems.

The core feature that manages this is Country/Region-Based Invoice Rules. When a supplier submits an invoice, SBN detects the originating country based on the supplier’s VAT ID prefix or address, then applies a specific set of validation rules that override default transaction rules. This ensures that suppliers in different countries — for example Italy vs the United States — see different required fields and localized behaviors.

The platform supports several compliance models depending on the region.

For VAT-regulated regions, particularly in the EU, Australia, and Canada, SBN enforces required tax information such as VAT IDs, supply dates, tax breakdowns, and local currency requirements.

In countries using government clearance models, where the government must approve an invoice before the buyer receives it, SBN integrates directly with national invoice platforms. Examples include KSeF (Poland), SDI (Italy), and the E-Invoice system (India).

SBN automatically applies digital signatures to invoices for many countries to ensure document integrity and authenticity.

The system also converts standard cXML invoice data into legally mandated formats required by specific governments or networks. Examples include FatturaPA (Italy) and Peppol UBL (supported in countries such as Belgium and Singapore).

Buyers interact with these features primarily through configuration rather than manual invoice transactions.

To configure rules, buyers navigate to Country-Based Invoice Rules within the Administration and Configuration settings. Here, countries can be activated and compliance requirements can be enabled or disabled, such as requiring suppliers to send tax invoice documents.

Buyers can monitor invoice submission status in their inbox. For government-connected countries, invoice history will include status updates showing government validation progress and results.

Buyers can also generate compliance visibility through built-in reporting. One example is the Supplier Legal Profile report, which shows which suppliers have completed the required tax compliance setup for their country or region.

Suppliers see a dynamic invoice creation experience based on their country. During invoice creation, suppliers are shown required local tax fields based on their location — for example GST IDs and HSN codes in India, or Stamp Duty fields in Italy.

Suppliers in countries with strict mandates must complete a Legal Profile or Country Configuration in their account settings. This is where they enter tax registration numbers and upload any required digital certificates or identifiers for government or network compliance.

To enable these features in an SBN Buyer account, buyers must update Default Transaction Rules to allow country-based invoice rules to take precedence instead of being ignored.

Buyers must also add and activate the specific countries they trade with in Country-Based Invoice Rules.

For countries that route invoices through external networks like Peppol, buyers must register their Peppol Participant ID or equivalent identifier in their profile so suppliers can send invoices successfully.

For countries where SBN retrieves invoices directly from government portals (inbound invoice flow), buyers must configure External Document Service settings to establish a connection with the national tax authority. Examples include Poland and Romania.

SBN supports country-specific compliance logic, VAT validation, or government integration for the following countries.

Americas: Brazil, Canada, Chile, Colombia, Mexico, Peru, United States.

Asia/Pacific: Australia, China, India, Japan, Malaysia, New Zealand, Singapore, South Korea, Taiwan, Thailand.

Europe: Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Liechtenstein, Lithuania, Luxembourg, Netherlands, Norway, Poland, Portugal, Romania, Russia, Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey, Ukraine, United Kingdom.

Middle East/Africa: Israel, Saudi Arabia, South Africa, UAE.

Considerations & Final Takeaways

Every organization wants onboarding to be faster, clearer, and built on trusted knowledge rather than fragmented internal know-how. NotebookLM enables teams to convert authoritative product documentation, SOPs, configuration specs, SOWs, demos, and project artifacts into a secure AI-powered conversational resource that employees can interact with naturally—like chatting with their most knowledgeable coworker. Instead of manually searching across files and web pages, teams can ask direct questions about product capabilities, setup requirements, and configuration considerations, then instantly convert those answers into internal enablement assets such as training outlines, FAQs, flashcards, infographics, and learning summaries. When setting up your NotebookLM hub, prioritize source quality, clear document labeling, and intentional prompting during training to reduce ambiguity and improve accuracy. This approach scales product fluency across teams, streamlines learning without adding process overhead, and creates more confident, self-sufficient users—while ensuring that expert conversations, when needed, are informed, efficient, and outcome-driven.

From Hype to Hands-On: Level Up Your AI Skills This January

2026-01-16T09:46:40.184000+01:00

In the age of AI, the skills we build today will define how we work tomorrow. That’s why we’ve launched a monthly series of blogs that spotlight the latest AI learning assets and upcoming learning opportunities. This blog explores the latest learning and enablement resources on SAP Document AI, as well as the bi-directional integration between Joule and Microsoft 365 Copilot. We also dive into some key resources to help you get started building your own Joule agents.

Building on my first blog, where I introduced you to our AI live sessions Beyond Hype (which is all about getting productive with SAP Business AI), I wanted to address some of your most asked questions:

How can I enable SAP Document AI on my mobile device?
SAP Document AI also supports mobile use cases, ideally for employees working remotely or in the field who need to submit documents while on the go. Within SAP Mobile Start, you can configure Document AI as a target, then take a picture and automatically upload it for processing. Find out here how to integrate and launch SAP Document AI on mobile. If you haven’t set up SAP Document AI yet, you want to check out the steps here.
Did you know about the recent feature release: Inbound and outbound channel support, which allows you to connect your e-mail server, as many document-related use cases involve e-mail. For now, only Microsoft Outlook is supported, but this enables you to fetch the e-mail, extract both the e-mail body and attachments, such as Excel files. For e-mail ingestion, you don’t need BPA integration; you can also use our e-mail ingestion feature without BPA by configuring your e-mail inbox as an inbound channel in our UI. Read more about the Q4/25 Release Highlights here and follow the documentation to register an e-mail app here.
Stay up to date with SAP Road Map Explorer and learn how you will benefit from SAP Document AI in upcoming releases.
Where can we find instructions for setting up the bi-directional integration between Joule and Microsoft 365 Copilot?
The great thing is that you can get started today with Joule Base and a Microsoft 365 license by downloading the SAP Joule app from the Microsoft Marketplace (M365 Copilot license required when using M365 Copilot skills in Joule web client in your SAP Application). Then follow the SAP Help documentation to set up the connection between Cloud Identity Services on SAP BTP & Microsoft Entra, and consider a few technical pre-requisites:
1. Is Joule activated in at least one SAP Business Applications (e.g., SAP SuccessFactors or SAP S/4HANA Cloud), and do you have administrator access to SAP Cloud Identity Services, SAP Business Technology Platform (BTP), and Microsoft Entra? Yes? Great, read on!
2. Next, you need to:
- Configure Microsoft Entra as your Corporate Identity Provider
- Enable Joule to Pass Requests to Microsoft Copilot and Teams
- Joule Formation to be included with MS Copilot Integration in your system landscape
Watch my colleague @nageshcaparthy demo the setup process!
May we use Joule within Microsoft 365 Copilot (and vice versa) when handling confidential data?
Did you know that Joule and Copilot are only exchanging the user prompt and displaying the result in their respective UIs? Through Identity Propagation between Microsoft Entra and Cloud Identity Services on SAP BTP, Joule and Microsoft 365 Copilot are aware of a user’s roles and permissions for a secure integration. Watch the video and see the two Copilots work together in action!

Want to dive deeper into the content? Rewatch the sessions and access the session material via SAP Learning Hub (subscription required):

Getting Started with SAP Document AI: Activate, Automate, Act (recording)
Two Copilots, One Goal: Smarter Productivity with Joule and Microsoft 365 Copilot (recording)

Speaking of automation and productivity, have you heard that Agent builder in Joule Studio is now generally available? The agent builder enables developers to create Joule agents tailored to their unique business requirements, breaking cross-ecosystem silos and reducing barriers to AI adoption. Read on in Michael Ameling’s Blog, President of SAP Business Technology Platform, to explore key highlights of the release, everyday use cases for Joule agents, and get insights into the roadmap.

Here are your key resources to help you get started building your own Joule agents:

How to Get Started with Joule Studio: Prerequisites, Commercials, and Instructions to begin using Joule Studio
Joule Studio Goes Live: Answers to Your Top Questions covering Skills and the new Agent Builder features
Create an Agent in Joule Studio: Step-by-step tutorial on how to build and test an agent in Joule Studio
Best Practices for Building Agents: Key principles and practices for designing effective Joule Agents
Discovery Center Mission: Get Hands-on experience in creating a Joule Agent with multiple skills and MCP

Hi Joule, where should I get started with the latest learning resources? *

Apart from live enablement sessions, explore our latest on-demand learning resources:

To all partners: Test, demo, and develop with Joule and Microsoft 365 Copilot by purchasing the Joule Base TDD entitlement (8020592) and any Microsoft 365 environment (you don't necessarily need M365 Copilot to get started). Check out the Microsoft 365 Developer Program to get started.
Get hands-on with SAP Document AI! Pre-configured content is automatically delivered as part of the application and can be activated in the schema UI or via the schema APIs. In the video, you will learn how to create your own schemas from scratch in the SAP Document AI workspace. Now it’s your turn - Go ahead and try it yourself!

* Powered by the machine learning technology of SAP AI Business Services, personalized recommendations are generated for employees, in one place on the Learning home page, based on employee data.

As we look ahead, we’re excited to share with you a new lineup of learning opportunities designed to help you expand your knowledge and make AI an advantage in your role:

February 5: What’s New in SAP Build Q4 2025 Option 1 9 am CET - Option 2 4 pm CET
February 10, 5 pm CET: Reimagining Business Processes with Joule Agents. Discover how Joule Agents, embedded in core business processes, help organizations transition from insight to action.

Stay tuned for more, and happy learning! 🚀

Further Resources:

Joule Joule Studio SAP Document AI #Microsoft 365 Copilot SAP Mobile Services

Stop Building from Scratch: Autonomous AI Agents for SAP Sales Order Processing

2026-02-07T18:50:48.664000+01:00

Why this article?

It is great that we are living in the era of AI technologies where humans are still needed to design and automate 😀. But let’s be honest: give it a few more years and no company will ask "Which AI platform should we choose?" That decision will be made by central AI services based on strict country and corporate regulations. The only thing that will matter is security and reliability.

So, why am I writing this? Because I see so many people struggling with custom, built-from-scratch AI automation solutions at the enterprise level. These projects start quickly, but they get stuck just as fast.

You know, it is like cookies 😀. You can make one in a sandbox; however, to scale that to a corporate production level is a totally different problem.

If you are still wondering why this article matters: I want to highlight that while humans are still designing automation, speed is everything. Deploying autonomous AI agents on a solid, reliable platform is a massive competitive advantage. Those playing around with internal IT projects or relying on System Integrators (SI) to build from zero are being left behind.

Let me explain why using one simple example: Customer Purchase Orders.

The Business Case: The Reality of POs

A company wins a deal (congratulations!), and the Customer sends a Purchase Order (PO).

In an ideal world, this happens via EDI. If you have EDI, you are lucky and you can stop reading. But in the real world, customers send PDFs, scanned images with handwriting, Excel files with macros, or even a screenshot of their laptop screen.

This shouldn't be an issue. Innovative platforms must handle data extraction from unstructured sources out of the box.

“Training? Fine-tuning??”

I know some of you just thought that 😊. Yes, training OCR models used to be mandatory. If you haven't faced the pain of training models, be happy, because you are young enough 😊. Today, this must work without that hassle.

The Main Challenge: Master Data Mapping

Extracting text is easy. The real challenge is my favorite one: Master Data Mapping.

How do we robustly and reliably read the customer’s product description and map it to the correct Master Data Object in our SAP system? And how do we do it without heavy development?

The Solution: SAP Master Data Vectorization

We don't need custom code; we need Vectorization.

Generative AI can write SQL queries, sure. But think about it: how many SQL queries would the AI have to guess to find a material that is semantically similar but written differently? Infinite. Or maybe a few less if you are lucky and the description contains an acronym like "A4 Paper" 😊.

When AI looks for a material master in vectorized data (rather than a database table), it reads by sense, not just by keywords. This allows it to reliably find the right SAP object even if the customer's description is vague.

I know only one SAP-certified Business AI Platform that has this Vectorization feature in its standard set: Skybuffer AI.

Pic.1 SAP Master Data Vectorization Setup

Pic.2 Example of Material Master Similarities based Search

The Implementation: Autonomous Agents in Action

This is the essential part: stop playing in the sandbox. Take a platform and create a simple autonomous AI agent.

In this example, I am using Skybuffer AI because it handles the heavy lifting out of the box:

Vectorizes SAP master data for semantic search.
Reads Emails (including complex attachments and OCR).
Safety First: Uses Mailboxes as a staging area (no messy data replication).
ABAP Control: It doesn't let the AI "guess" how to create data. It passes structured JSON to an ABAP class/BAPI for safe execution.
No Code: configured entirely in SAP Fiori.

The Flow

A Skybuffer AI agent acts effectively as an automated employee. You can schedule it, give it tools, and let it run.

Pic.3 Scheduling Autonomous AI Agent

Step 1: The Input

The Agent connects to a Mailbox. It reads the email, extracts the attachment data, and places it into the Agent's memory.

Pic.4 Flow inside the Tool of Skybuffer AI Action Server

Step 2: The Brain (Generative AI)

We connect the Generative AI and give it guidelines.

Note: The vectorized SAP master data is mapped directly to this action.

Pic.5 Generative AI Action of Skybuffer AI Action Server

The AI creates a JSON structure based on our request. Crucially, we do not ask the AI to update SAP directly. We hand that JSON over to an SAP ABAP class. This ensures the transactional data is created correctly, without hallucinations.

Step 3: The Output

We update the email in the Mailbox with the status and send a confirmation back to the customer.

Everything runs either in your SAP HANA On-Premise shell (directly in your network) or on your SAP BTP tenant.

Conclusion

Autonomous AI agents are no longer "future tech", they are easy solutions available now.

By choosing a Business AI automation platform rather than building from scratch, your company gets the best of Generative AI and the SAP backend immediately.

It is strange that so many companies are still "playing" with in-house development. You wouldn't build your own Sales Order Management system from scratch instead of using SAP, right? 😊 That would be weird. Yet, companies are giving huge budgets to internal teams to build AI wrappers.

You don't need a giant team. You need a reliable solution, speed, and flexibility. Just look at Cursor, they revolutionized coding with roughly 60 employees before their massive 900M funding round 😉.

Be quick. Be autonomous.

Efficient Document Inbound Processing in SAP Cloud ERP with SAP Document AI: March 12, 2026

2026-02-09T21:12:08.498000+01:00

Managing incoming documents efficiently is critical for business operations. SAP Cloud ERP, powered by SAP Document AI, offers pre-delivered use cases to automate and optimize document processing across Sales, Finance, and Quality Management.

In this live session, we’ll explore how SAP Document AI enhances document inbound processing, reducing manual effort and improving accuracy. You’ll see live demos of end-to-end workflows and learn how to activate these capabilities in your system.

Key Takeaways:

Discover pre-built document processing use cases in SAP Cloud ERP.
See live demos of end-to-end workflows for Sales, Finance, and Quality Management.
Learn how to activate and extend these capabilities for your business needs.

Who Should Attend? Project managers, consultants, business users, and administrators looking to improve document processing efficiency.

Register Now!

Leveraging SAP Build Process Automation for Intelligent Invoice Extraction from Outlook

2026-02-17T08:27:43.106000+01:00

Imagine a world where your inbox doesn’t just store invoices, but actively processes them. No more manual downloads, no more mind-numbing data entry, and zero "human-error" bottlenecks.By integrating the "digital eyes" of SAP Document Information Extraction with the orchestration power of SAP Build, organizations can now transform unstructured email attachments into clean, high-fidelity structured data in seconds. This blog explores the seamless end-to-end journey of an invoice—starting from the moment it hits an Outlook inbox, through its automated recognition and validation phase, to its final state of readiness for your core business systems.

Phase 1: Environment Readiness

The Local Bridge: The Desktop Agent 3.0 must be installed and registered to the tenant. This acts as the execution engine, enabling the cloud-based SAP Build environment to interact directly with the local Outlook application and file system.

-To set up the agent, follow this: Install and Set Up Desktop Agent 3 | SAP Tutorials

Phase 2: Agent Configuration: Creation, Assignment, and Project Binding

Once the agent is installed, it must be assigned to an environment within the SAP Build Lobby to authorize automation execution.

Create the Environment: Navigate to Control Tower > Environments and select Create. Provide a name, Identifier and click Create.

Register the Agent: Open the newly created environment and select Agent Management > Add Agent.

Define Agent Attributes: Navigate to Control Tower > Agent Attributes and select Create. Provide a name and choose the Type:

Free: For dynamic parameterization of values.

Predefined: For assigning specific, fixed values.

Assign Attribute to Agent: Under Control Tower > Agents, locate your agent and ensure it is connected. Select Manage Attributes, choose your newly created attribute, enter the corresponding value, and confirm.

Bind Attribute to Project: In your project’s Process Builder, navigate to Settings > Attributes. Add the same attribute and value, then save. This creates the final link, ensuring the project triggers specifically on your configured agent.

Phase 3: Building the Automation Logic

Now that the technical "handshake" is configured, we can begin building the automation. This phase focuses on creating a reliable bridge between your email inbox and your local storage.

Initializing the Automation

Start by creating a new automation in the Process Builder. This will be the engine that handles your file and email operations. Ensure you have the Outlook SDK and File System SDK added as dependencies in your project settings.

The following is the flow for extracting info from invoices, which includes opening Outlook, searching for specific emails, saving attachments, and processing the data:

Phase 4:Preparing the Workspace: Local and Outlook Setup

For the automation to successfully process documents, you must establish a specific "landing zone" for your files. This requires the following manual configurations:

1.Local File System: Manually create a folder on your computer at the path C:/temp/attachments. This folder acts as the dedicated directory where the bot will download and store email attachments before they are uploaded to the cloud or processed by the extraction logic.

2.Outlook Folder Configuration: Inside your Outlook application, you must create a new folder specifically named Invoices.

Setup for Testing: To test the flow, move an email containing an invoice attachment into this Invoices folder.
The Unread Status: Ensure the email is marked as Unread. The automation is designed to search for unread items to prevent processing the same invoice multiple times.

Phase 5: Configuring BTP Destinations

Destinations are secure connection points that allow your automation to communicate with external services. This is a two-step process: first creating them in the BTP Cockpit and then configuring them within SAP Build.

Creation in SAP BTP Cockpit

In your BTP Subaccount, navigate to Connectivity > Destinations and create the following:

1.Service & Extraction Destination: Create a destination to handle both internal routing for the Process Automation engine and the connection to the Document Information Extraction (DOX) service.

2.SAP Document Store Destination: Create a mandatory destination for sap_process_automation_document_store. This allows the bot to move files from your local C:/temp/attachments folder into the secure SAP Document Management Service (DMS) for centralized cloud archiving.

Configuration under SAP Build

After creating them in BTP, you must explicitly link them to your project to establish the necessary communication channels:

Mapping: In your project settings, go to the Destinations tab. Map your BTP destinations to the corresponding project requirements. This step ensures that the automation knows exactly which BTP service to call for extraction and storage.

Phase 6: Deployment and Execution

With the logic built, folders prepared, and destinations mapped, the system is now a cohesive unit. The final step is to move the project from the design stage to active execution.

1.Releasing and Deploying

In the SAP Build Lobby, you must Release your project to create a version snapshot. Once released, navigate to the Deploy button. This sends the instructions to the cloud environment, where it waits for your local agent to "pick up" the task.

2.Monitoring

Once the automation completes, navigate to the Monitoring tab in SAP Build. Here, you can view the execution logs for your job.

grossAmount: The bot has correctly identified the total (e.g., $93.50).

filePath: The bot confirms where it stored the document locally (e.g., C:\temp\invoice-a (1).pdf).

recipient: It captures the specific email address the invoice was sent to.

3.The Final Flow in Action

The following is the automated flow created for extracting info from invoices:

Outlook Monitoring: The bot identifies an unread email in your manually created Invoices folder.

Local Storage: It downloads the attachment directly into C:/temp/attachments.

Data Extraction: Once the file is stored locally, the bot uses the extraction activity to communicate with the cloud. At this stage, we can successfully obtain the extracted info from the invoice.

Conclusion:

By linking Outlook to SAP Document Information Extraction through SAP Build, you transform the inbox from a storage bin into a high-speed data engine. This setup doesn't just save time—it eliminates manual errors and ensures your business systems receive accurate data instantly. It’s a simple yet powerful blueprint for turning administrative friction into a seamless, automated advantage.

Disclaimer:
All the views and opinions in this blog are my own and are made in my personal capacity. SAP shall not be responsible or liable for any of the contents published in this blog.

SAP BTP AI best practices - SAP Document AI

2026-02-17T09:00:00.015000+01:00

Still extracting document data manually? Let’s change that.

In our next SAP BTP AI best practices webinar, we spotlight SAP Document AI — and how it automates extraction of structured and unstructured information on SAP Business Technology Platform.

Invoices. Contracts. Forms. Done.
Machine learning + generative AI + instant learning. Even better.

This isn’t theory. It’s practical guidance you can use immediately — in customer demos, project delivery, or new AI proposals. The opportunity is real, and partners who master document intelligence early will stand out.

Short session. High impact. Real use cases.

Secure your spot. Register now. https://partneredge.sap.com/en/library/education/psd/2025/nov/e_oe_te_w_PSD_WEB_00012198.html

#SAP #SAPBTP #SAPDocumentAI #AIUseCases #SAPPartners #Innovation

SAP Community - Document Information Extraction

New in SAP Mobile Start V2.4: Integrate and Launch SAP Document AI on Mobile

Preconditions

Subscribe to SAP Document AI

Enable Identity Authentication for SAP Build Work Zone

Configure Trust Relationship for API Access

Create Destination for SAP Mobile Start

Configuring tile for SAP Document AI access in mobile

Step 1: Create a New App in Site Manager

Step 2: Configure Navigation Parameters

Step 3: Configure Visualization

User Experience in SAP Mobile Start

Upload Options

Upload Validation & Handling

Track Status and Review Document

Conclusion

Instantly improve document information extraction

Using Document AI Premium in SAP Build Process Automation

Intro

Prerequisites

Easiest way to start

Notes for advanced automation users

Conclusion

Leveraging SAP Document AI to bring intelligent document processing into a classroom

New Expert-Guided Implementation: Unlocking Joule and SAP Business AI from Strategy to Activation

Unlocking AI: Joule and SAP Business AI from Strategy to Activation

Workshop Schedule

What You’ll Learn

Why Attend

How to Register

Related Articles

Integrating SAP Document Information Extraction (Document AI) with SAP CAP – Step by Step

New Expert-Guided Implementation: Unlocking Joule and SAP Business AI from Strategy to Activation

Unlocking AI: Joule and SAP Business AI from Strategy to Activation

Workshop Schedule

What You’ll Learn

Why Attend

How to Register

Register here to secure your spot* today to learn from SAP experts and gain the skills to bring Business AI to life across your enterprise.

[2025-Q4] SAP BTP Adoption Webinars – Unlock the Power of AI, Data, and Integration

📅 Oct 30 | Pipeline Concept in SAP Integration Suite

📅Nov 04 | Service and Service Excellence with AI Driven Insights

📅 Nov 05 | Best practices for secure and scalable LLM model usage in BTP AI apps with SAP AI Core

📅 Nov 13 | AI-Driven Business Solutions – Exploring the Potential of SAP BTP AI - Americas

📅 Nov 13 | Enablement on SAP Mobile Start and SAP Build Work Zone with Showcase of Joule as AI Copilot

📅 Nov 19 | AI-Driven Business Solutions – Exploring the Potential of SAP BTP AI - APAC/EMEA

📅 Nov 20 | BTP AI Best Practices – Knowledge Graph-based Retrieval Augmented Generation

📅 Dec 03 | BTP AI Best Practices – Vector-based RAG, Document Grounding, and Examples

📅 Dec 09 | AI-Driven Business Solutions – Exploring the Potential of SAP BTP AI (Americas Session)

📅 Dec 15 | AI-Driven Business Solutions – Exploring the Potential of SAP BTP AI (APAC/EMEA Session)

SAP Document AI in SAP MDG using Intelligent Scenario Lifecycle Management (ISLM)

Joule - Document Grounding with SAP Build Work Zone, advanced edition (RAG)

Setting up Document Grounding for SAP Build Work Zone, advanced edition

Document Grounding now supports two authentication mechanisms. Please check the Configure Access from SAP BTP before you start your setup.

Setup Steps

Step 1: Enable Document Grounding capabilities in SAP Build Work Zone, advanced Edition

Step 2: Enable Document Grounding for Workspaces & Sub-Workspaces in SAP Build Work Zone, advanced Edition

Step 3: Explore Content Locations — Blogs, Wikis, and Documents for Grounding

Step 4: Use an Existing Joule Subaccount or Create a New One to Set Up Document Grounding

4.a Scenarios for Setting Up Document Grounding

4.a.1 – Unified Joule Already Exists with SAP Build Work Zone, standard Edition

4.a.2 – Unified Joule Already Exists with SAP Build Work Zone, advanced edition

4.a.3 – Setting Up Unified Joule for the First Time and Connecting an Existing SAP Build Work Zone, advanced edition

4.a.4 – Setting Up Unified Joule for the First Time (SAP Build Work Zone, advanced edition in the Same Joule Subaccount)

Optional Steps

After completion of Step 4.a.1 or 4.a.2 or 4.a.3 or 4.a.4 proceed to create a Destination

Step 5: Create Certificate and RSA Key Files

Step 6: Run cURL commands to setup Document Grounding pipelines

6.a Install Bruno and Create a Collection

6.b Create a Bearer Token

6.c Create a Pipeline

Setup Complete

Step 7: Verify and Test Joule

Step 8: Appendix

8.a. Get All Pipelines

8.b. Delete a Pipeline

8.c. Trigger a Pipeline

8.d. Check Pipeline Execution/Grounding Status of the documents

8.e. Check Pipeline Status & Total executions

SAP Document AI accelerates document-centric business processes with Generative AI