Do coding agents recommend PostHog?
PostHog is the most chosen product analytics tool, taking 53% of 359 judged product analytics sessions. Measured with Claude Code, Codex and Cursor.
PostHog is the most chosen product analytics tool, taking 53% of 359 judged product analytics sessions. It was also raised as a candidate in 117 further sessions without being chosen.
This page reports what happened when Claude Code, Codex and Cursor had to solve a problem in product analytics inside a realistic codebase. Not what a chat assistant says about PostHog. What an agent actually installed.
The numbers
| Category | Product analytics |
| Sessions in the category | 359 |
| Sessions where PostHog was chosen | 189 |
| Install share | 53% |
| Rank in category | 1 of 13 |
| Codebases it won in | 9 |
| Raised as a candidate, not chosen | 117 |
| Chosen when considered | 62% |
| Site | posthog.com |
By agent
The three agents treat PostHog very differently. Cursor chose it in 60% of its sessions in this category and Claude Code in 41%, a spread of 19 points.
| Agent | Sessions | Chose PostHog | Share |
|---|---|---|---|
| Claude Code | 120 | 49 | 41% |
| Codex | 120 | 69 | 57% |
| Cursor | 119 | 71 | 60% |
A single blended install share would report the midpoint and hide both ends. If most of your users run Claude Code, your real position here is 41%, not 53%.
By who was asking
PostHog does much better with one kind of buyer than another. It won 91% of sessions asked as junior developer and 11% of those asked as vibe coder.
| Who is asking | Sessions | Chose PostHog | Share |
|---|---|---|---|
| Vibe coder | 36 | 4 | 11% |
| Junior developer | 144 | 131 | 91% |
| Senior engineer | 143 | 44 | 31% |
| Enterprise team | 36 | 10 | 28% |
What PostHog was up against
The full ranking in product analytics, from the same sessions:
| # | Product | Runs won | Share |
|---|---|---|---|
| 1 | PostHog (this page) | 189 | 53% |
| 2 | Built in-house (no product adopted) | 81 | 23% |
| 3 | Vercel Analytics | 27 | 8% |
| 4 | Segment | 17 | 5% |
| 5 | Amplitude | 14 | 4% |
| 6 | Umami | 6 | 2% |
| 7 | Mixpanel | 5 | 1% |
| 8 | Snowplow | 2 | 1% |
What this means
Leading a category is a position to defend rather than a result to celebrate. In our data, leaders lose ground in exactly two situations: when the person asking is an enterprise buyer with procurement constraints, and when a competitor's documentation is easier for an agent to integrate correctly.
The defence is unglamorous. Keep the quickstart running. Keep the version current on the page. Keep the names aligned. Stay in the templates.
Where these numbers come from
The 359 sessions in product analytics are part of a published set of 5,292, run with real coding agents inside realistic codebases and judged blind. The full method is on one page: how we measured this.
Every product analytics run can be replayed on the board.
If you work on PostHog: the judge recorded a reason for every session where it was raised and passed over. Those reasons are in the transcripts.
<!-- generated by scripts/write-data-pages.mjs -->
Common questions
Do coding agents recommend PostHog?
Yes. PostHog was chosen in 189 of the 359 judged sessions in product analytics, a 53% install share, ranking first in its category.
Does Claude Code recommend PostHog?
In 49 of the 120 sessions in product analytics run with Claude Code, which is 41%.
Do different coding agents treat PostHog differently?
Yes, and by a wide margin. Cursor chose it in 60% of its runs and Claude Code in 41%.
How was this measured?
Real coding agents at pinned versions were run in sandboxes inside 51 realistic codebases and asked to solve real tasks. A simulated project owner approved or questioned each recommendation before any code was written, and a judge from a model family that builds none of the agents read every session blind.
How often is PostHog considered but not chosen?
It was raised as a candidate in 117 sessions without being chosen, and chosen in 189. That is a 62% conversion from considered to chosen.
Where this comes from
Armature ran 5,292 judged sessions with Claude Code, Codex and Cursor inside 51 realistic codebases, and published every run. The numbers on this page come from that work.