AI Adoption Score vs Claude Code, Codex and GitHub Copilot analytics

Claude Code, Codex and GitHub Copilot each come with analytics for their own tool: how many people use it, how much of its code they accept and what it costs, for the people on your company's plan. AI Adoption Score asks people how well they use AI, whichever tools they use, and what holds back the ones who don't. The two fit together.

Last checked: October 17, 2026

Which one fits you

Two questions decide it.

Side by side

Scores out of 5: 4–5 strong 3 partly 1–2 weak or missing

Question	AI Adoption Score	Vendor analytics
What it measures	Score: 5 out of 5.How people adopt AI: a level for every person and team across usage, skills, impact, culture and vision (opens in a new tab)	Score: 4 out of 5.Use of each tool: active users, sessions or messages, code accepted, pull requests and code reviews, and spend; Claude Code and Copilot add estimates of value and cost^{1, 2, 7, 13}
AI tools covered	Score: 5 out of 5.Any: Copilot, ChatGPT, Claude, Gemini, Cursor, and tools people pay for themselves	Score: 2 out of 5.One each: Claude Code's analytics count Claude Code, Codex's count Codex and Copilot's count Copilot^{1, 6, 11}
Whose use it counts	Score: 5 out of 5.Everyone you invite, about every AI tool they use, including the ones they pay for themselves	Score: 3 out of 5.People on your company's plan: Claude's usage analytics come with Team and Enterprise, not Pro or Max; Codex's with ChatGPT Enterprise and Edu, not Business; Copilot counts people with a seat^{3, 8, 11}
Where the data comes from	Score: 3 out of 5.Surveys and short quizzes that people answer	Score: 4 out of 5.Recorded by each tool as people use it^{1, 7, 11}
Measured AI coding use	Score: 3 out of 5.AI Coding Usage: sessions, repositories, tools and connectors from Claude Code and Codex in one report, for each engineer who connects their Mac (opens in a new tab)	Score: 4 out of 5.Exact use of their own tool, down to the lines people accept and the pull requests they merge; Copilot sorts people into adoption phases^{1, 11, 12}
What you need	Score: 5 out of 5.Nothing for the surveys: a survey link, or invitations to an email list. For AI Coding Usage, each engineer pastes one line on their Mac	Score: 5 out of 5.Nothing new to buy: it comes with the plan; Claude's pull-request figures need its GitHub app, and Copilot's need a policy switched on and IDE telemetry^{1, 11, 13}
Why people don't use AI	Score: 5 out of 5.18 barriers, each with who can fix it: IT, leadership, managers or training (opens in a new tab)	Score: 2 out of 5.They show who uses the tool and how much, not why the rest don't; GitHub's own guide suggests asking people as well^{7, 15}
AI skills	Score: 4 out of 5.Who can do what with 125 AI tools and models, and quizzes that score n8n and Claude skills (opens in a new tab)	Score: 2 out of 5.Use, such as lines accepted or how people work with Copilot, not skill^{1, 12}
Attitudes	Score: 5 out of 5.AI anxiety on a validated research scale, plus culture and vision in the 360° (opens in a new tab)	Score: 2 out of 5.Claude Enterprise can send a short survey in chat, in beta; GitHub suggests short pulse surveys of your own^{4, 15}
Levels	Score: 5 out of 5.L1 Observer to L5 Native for every person, team and the organization	Score: 2 out of 5.Copilot's four adoption phases, from Passive to Multi-agent; no maturity levels¹²
Privacy	Score: 4 out of 5.Named or anonymous, set per survey; results from the first answer, with no minimum group size; AI Coding Usage never sends prompts, code or file names	Score: 3 out of 5.Use per person for admins: a leaderboard and an export of everyone in Claude's, and user-level reports in Copilot's API^{1, 16}
Price	Score: 5 out of 5.Free plan; paid plans on request	Score: 5 out of 5.Included with each tool's business plans: Claude Team and Enterprise, ChatGPT Enterprise and Edu, Copilot Business and Enterprise^{3, 8, 14}
Summary	Every AI tool and every team, and the reasons behind the numbers: levels, skills, barriers with who can fix them, and worries, with a free plan to start.	Exact use of each tool, included with its business plan. But one tool each, for the people on your plan, and they show who uses it, not how well or why the rest don't.

Where each one sits

AI Adoption Score Surveys and quizzes on how people adopt AI, in engineering and business roles: levels, skills, barriers, attitudes.
Coding tools' own analytics Claude Code, Codex and Copilot each count their own tool's use, for the people on your plan.

Positions are our reading of each product's public pages in October 2026, not a measurement.

Where their analytics are stronger

Exact use, without asking anyone. Each tool records its own use as it happens, down to the lines people accept and the pull requests they merge. Our surveys know what people tell us, with checks, and AI Coding Usage counts sessions.^{1, 11}
Already included. They come with the plans your company already pays for, so there's no new vendor to approve.^{3, 8, 14}
Code that shipped. Claude counts the merged pull requests with code from Claude Code, through its GitHub app, and Copilot counts pull requests created and merged, and the median time to merge. We don't read code.^{1, 11}
Cost next to use. Claude shows spend by model and exports it person by person, and Copilot's return-on-investment section sets the cost per developer against the pull requests per developer, for each adoption phase.^{2, 13}
APIs for your own dashboards. Claude Enterprise, ChatGPT Enterprise and Copilot each have an API for the same numbers.^{1, 9, 16}

Using both

The vendors say their numbers are a start. Anthropic designed its metrics to complement your engineering KPIs, GitHub asks you to treat its return figures as directional, and OpenAI says usage alone doesn't establish savings.^{5, 10, 13}

Use their analytics to see who uses each tool and how much. Then ask the people: the 360° AI Maturity Assessment (opens in a new tab) gives every person and team a level, AI Adoption Barriers (opens in a new tab) shows what stops the people who don't use AI and who can fix it, and the Claude Skills Assessment (opens in a new tab) shows how well people know Claude, Claude Code included.

With more than one tool, each dashboard shows a part. Our surveys ask about every AI tool, the ones people pay for themselves included, and AI Coding Usage (opens in a new tab) puts Claude Code and Codex in one report.

Questions

Do these analytics count people on personal plans?

No. They count the people on your company's plan. Claude's usage analytics come with Team and Enterprise, not Pro or Max, and cover the people in your Claude organization; Codex's come with ChatGPT Enterprise and Edu, and cover your workspace. Someone on a personal subscription doesn't show up in either.^{1, 3, 8, 9}

We have Copilot. Does AI Adoption Score read its metrics?

No. It connects to nothing in GitHub or anywhere else. Read the two side by side: Copilot's numbers for use, our reports for the reasons behind them.

Does AI Adoption Score measure Claude Code and Codex?

Yes. AI Coding Usage (opens in a new tab) counts sessions, repositories, tools and connectors from Claude Code and Codex in one report, for each engineer who connects their Mac with one line. Prompts, code and file names never leave the Mac. It works on Macs for now, and it doesn't count Copilot yet.

Sources

Dashboards count use. Ask why the rest don't.

See how well people use every AI tool, the ones they pay for themselves included,
and what holds the rest back, in every team.

Start Free Book a Demo