---
layout: page
title: "Sentinel: Your Web-Performance Watchman"
page-class: page--services  page--sentinel
meta: "I keep an eye on site-speed so you don’t have to"
main: "https://csswizardry.com/img/content/sentinel/speedcurve-003.png"
permalink: /sentinel/
lux: "Services / Sentinel"
---

**I focus on site-speed so you don’t have to.**

Keep getting hit with performance regressions? Site-speed unstable and
inconsistent? No time to triage performance issues? No one taking ownership of
site-speed? **Sound familiar?**

Even the most performance-conscious teams rarely have the time or resources to:

* **define** appropriate performance budgets;
* **procure** and set up monitoring software;
* **identify** regressions;
* **separate** signal from noise;
* **triage** performance issues.

**That’s why they use _Sentinel_.**

Outsource your performance monitoring to me, your web-performance watchman.

<figure>
  <img src="{{ site.cloudinary }}/img/content/sentinel/speedcurve-001.png" alt="" width="1500" height="864">
</figure>

## What is Sentinel?

- - -

**sentinel** /ˈsɛntɪnəl/  
<small><i>noun</i></small>  
A soldier or guard whose job is to stand and keep watch.

- - -

Most companies simply don’t have dedicated performance teams (or even
individuals) to keep a constant eye on site-speed. Engineers have product and
feature work to do, and even with the best will in the world, performance ends
up falling off the radar. Regressions sneak in, go unnoticed, and never get
fixed.

<blockquote class="pull-quote  pull-quote--context-alt" id="quote:patrick-meenan">
<p><q>Service</q> undersells what Harry is providing here. It looks like a comprehensive monitor/triage/root-cause/impact high-touch service from one of the best.</p>
<b class="source  pull-quote__source"><a href="https://twitter.com/patmeenan/status/1648293161872117762">Patrick Meenan</a></b>
</blockquote>

_Sentinel_ is a service I offer to a small number of trusted clients: I become
your eyes and ears so you can get on with what you do best. Using my own
monitoring, I’ll work with you to define performance budgets and set up
relevant dashboards and alerts to make sure you stay fast.

The moment a regression is detected, I receive the alert and get to work. Your
developers don’t get interrupted by emails from monitoring software, and they
don’t need to wade though meaningless data to uncover meaningful insights. In
fact, they don’t need to do anything at all—that’s my job.

I’ll jump in, honing right in on the issue using bespoke dashboards and
performance budgets designed specifically for you. I’ll identify the problem and
hand over a report containing the timing of the regression, its cause, its
severity, its impact, and its fix.

<figure>
  <img src="{{ site.cloudinary }}/img/content/sentinel/speedcurve-002.png" alt="" width="1500" height="864">
</figure>

<p>
  <a href="https://calendly.com/csswizardry/sentinel" class="btn btn--full">Learn more…</a>
</p>

## What Does Sentinel Look Like?

There are three key aspects to _Sentinel_.

### Setup

Firstly, I’ll conduct an initial audit of your site to identify and address any
current slowdowns—there’s no point monitoring a slow website!

Then, we work together to outline the key metrics we’d like to track. Want to
monitor Largest Contentful Paint? Sure! TTFB? You got it! Total Blocking Time?
Done! Once we have a handful of metrics we’d like to keep an eye on, we gather
key pages to monitor—we could monitor all metrics on all pages, or any
permutation of the above. It’s entirely up to you.

Once we have our key metrics, it’s time to set some budgets. Most teams find
performance budgets paralysing, but that’s not your job anymore—it’s mine! I’ll
calculate relevant budgets for each of our chosen metrics, and set up the
corresponding dashboards and alerts for them.

<figure>
  <img src="{{ site.cloudinary }}/img/content/sentinel/speedcurve-003.png" alt="" width="1500" height="864">
</figure>

### Monitor

Now, we wait.

If any of our key metrics regress beyond our budgets, I get alerted immediately.
Alerts get sent to an expedited email address (that you also have access to),
shooting right to the top of my inbox.

From there, I begin analysing data, comparing charts, and running tests. I’ll
cross-reference many different vectors to hone in on the problem, working out
what happened and what might have caused it.

Once I have a reasonable idea of the issue and its solution, I’ll put together
a report and send it over. This could be via email, or directly into your issue
tracker. It’s entirely up to you.

<figure>
<img src="{{ site.cloudinary }}/img/content/sentinel/report-001.png" alt="" width="1500" height="864">
</figure>

### Report

<q>What if we don’t have any regressions?!</q>

This is great news! But don’t worry—you’ll never pay _Sentinel_ for nothing.
Regardless of how many incidents you experience, including zero, you’ll receive
the following reports every month and each quarter:

Monthly, you’ll receive a general performance report. How many incidents did we
have? How is the site faring compared to the previous period? Did the site
behave consistently throughout the month? Did we notice any patterns? Do we have
any opportunities outside of our key metrics?

Quarterly, you’ll receive a more comprehensive benchmarking report where we
compare your site against competitors and industry peers. How do we stack up? Is
anyone making great progress? Do we need to catch up to anyone? Can we identify
any opportunities?

<p>
  <a href="mailto:csswizardry@gmail.com" class="btn btn--full">Arrange a call</a>
</p>

## Pricing

**_Sentinel_ costs $5,999/mo.**

* <small>USD</small>
* <small>Billed quarterly</small>
* <small>20% discount for annual billing</small>
* <small>Minimum commitment of six months</small>

### What’s Included?

* **Initial [performance audit](/performance-audits/):** We’ll conduct
  a thorough audit before we begin monitoring. This allows us to start from
  a healthy baseline.
* **Metric and budget definition:** We’ll work out your key metrics and their
  corresponding budgets.
* **Monitoring software:** No need to procure your own monitoring software—I
  have it covered already! I’ll set you up in my SpeedCurve and Treo accounts.
* **Always-on monitoring:** The main event—I keep an eye on your performance
  budgets and tackle regressions as they happen.
* **Monthly status updates:** At the end of each month, I’ll summarise the
  previous period. Is site-speed trending up or down? Did we improve on last
  month? Did we have more incidents than normal?
* **Quarterly healthchecks and benchmarks:** Every three months, I’ll run
  a suite of tests against your closest competitors and peers to establish our
  position within your industry. Are we leading the pack? Or do we have
  opportunities for improvement?

## Learn More

If you want to know more about _Sentinel_ and how it can help your team, let’s
arrange a zero-obligation call to discuss it further:

<p>
  <a href="mailto:csswizardry@gmail.com" class="btn btn--full">Let’s talk <i>Sentinel</i></a>
</p>