---
id: ins_poyar-voje-context-engineering-gap
operator: Kyle Poyar
operator_role: Founder, Growth Unhinged newsletter
co_operators: ["Maja Voje"]
source_url: "https://knowledge.gtmstrategist.com/p/claude-for-gtm-pulse-report-2026"
source_type: research
source_title: Claude for GTM Pulse Report 2026
source_date: 2026-05-10
captured_date: 2026-05-10
domain: [pmm, ai-gtm, content-strategy]
lifecycle: [adoption, growth-loops]
maturity: applied
artifact_class: metric-model
score: { originality: 3, specificity: 5, evidence: 4, transferability: 5, source: 4 }
tier: A
related: [ins_context-engineering-beats-prompt-engineering, ins_maja-voje-substrate-first-content-engineering, ins_bottleneck-is-context-not-capability, ins_claude-plus-crm-plus-orchestration-stack]
raw_ref: 
---

# Among 200 GTM operators surveyed, detailed company context adoption separates systemic AI builders from task runners by 21 percentage points.

## Claim
Among 200 GTM operators surveyed, 93% of advanced Claude users have detailed company context built into their workflow as a foundation. Among task-runner users, 72% do the same. That 21-point gap is the measurable distance between teams reporting workflows they describe as previously impossible and teams running one-off prompts with generic output.

## Mechanism
A cold prompt gives the model no company-specific signal. It draws on generic training data and returns to the average of that data. When a shared context document carries ICP definitions, competitive claims, voice rules, and domain vocabulary, every query inherits that specificity. Output quality improves not because the model is smarter but because the input is more precise. Systemic builders maintain this context artifact and update it as the product or positioning shifts. Task runners rebuild context from scratch each session or never build it at all.

## Conditions
Holds when: a shared context artifact is maintained, accessible to the whole team, and updated when the product or positioning changes.
Fails when: context lives in one person's local files, goes stale after a product change, or is never connected to the team's prompting workflow.

## Evidence
37-page survey of 200 GTM operators across Pro, Max, Team, and Enterprise Claude plans. 67% of advanced users report workflows previously impossible. 55% have replaced a tool or vendor: ChatGPT, an agency, or a contractor.

> "The top GTM AI tool stack in 2026 is Claude + CRM + orchestration. These tools are not competing. They are combining."

## Signals
- Team members produce consistent, on-brand output without rewriting prompts each session
- New hires produce usable work in their first week by importing shared context
- Output drift triggers a context refresh rather than a model switch
- The team's AI-generated work is indistinguishable in voice and specificity from manually written work

## Counter-evidence
The survey skews toward self-selected Claude power users who opted into a GTM-focused research project. The correlation between context adoption and performance improvement may partly reflect selection bias: teams already invested in systematic work are more likely to build context artifacts and more likely to report strong outcomes. Base-rate adoption across the broader knowledge-worker population is likely lower.

## Cross-references
- ins_context-engineering-beats-prompt-engineering, Aatir Abdul Rauf
- ins_maja-voje-substrate-first-content-engineering, Maja Voje
- ins_bottleneck-is-context-not-capability, Sherwin Wu
- ins_claude-plus-crm-plus-orchestration-stack, Maja Voje