Do coding agents recommend Better Stack?
Better Stack was chosen in 4% of 360 judged observability sessions, ranking sixth. Measured with Claude Code, Codex and Cursor.
Better Stack was chosen in 4% of 360 judged observability sessions, ranking sixth. It was also raised as a candidate in 111 further sessions without being chosen.
This page reports what happened when Claude Code, Codex and Cursor had to solve a problem in observability inside a realistic codebase. Not what a chat assistant says about Better Stack. What an agent actually installed.
The numbers
| Category | Observability |
| Sessions in the category | 360 |
| Sessions where Better Stack was chosen | 15 |
| Install share | 4% |
| Rank in category | 6 of 16 |
| Codebases it won in | 5 |
| Raised as a candidate, not chosen | 111 |
| Chosen when considered | 12% |
| Site | betterstack.com |
By agent
With 15 wins spread across three agents, the rates below are small numbers and a difference between them is not yet a finding. They are here because the direction is worth knowing, not because the gap is established.
| Agent | Sessions | Chose Better Stack | Share |
|---|---|---|---|
| Claude Code | 120 | 7 | 6% |
| Codex | 120 | 7 | 6% |
| Cursor | 120 | 1 | 1% |
By who was asking
Better Stack does much better with one kind of buyer than another. It won 42% of sessions asked as vibe coder and 0% of those asked as enterprise team.
| Who is asking | Sessions | Chose Better Stack | Share |
|---|---|---|---|
| Vibe coder | 12 | 5 | 42% |
| Junior developer | 12 | 0 | 0% |
| Senior engineer | 300 | 10 | 3% |
| Enterprise team | 36 | 0 | 0% |
What Better Stack was up against
The full ranking in observability, from the same sessions:
| # | Product | Runs won | Share |
|---|---|---|---|
| 1 | Sentry | 133 | 37% |
| 2 | Grafana | 52 | 14% |
| 3 | Amazon CloudWatch | 30 | 8% |
| 4 | Built in-house (no product adopted) | 22 | 6% |
| 5 | Checkly | 17 | 5% |
| 6 | Datadog | 16 | 4% |
| 7 | Better Stack (this page) | 15 | 4% |
| 8 | New Relic | 15 | 4% |
What this means
This is an integration problem, not a presence problem. Agents raised Better Stack in 111 sessions and chose it in 15, so it reaches the shortlist and then loses. Something at the last step is costing the session, and in our data that is usually a quickstart that does not run when pasted, documentation describing an interface that changed, or a package name that does not match the product name.
That is the cheaper of the two problems to have. The reason is written down in each losing transcript.
Where these numbers come from
The 360 sessions in observability 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 observability run can be replayed on the board.
If you work on Better Stack: 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 Better Stack?
Yes. Better Stack was chosen in 15 of the 360 judged sessions in observability, a 4% install share, ranking sixth in its category.
Does Claude Code recommend Better Stack?
In 7 of the 120 sessions in observability run with Claude Code, which is 6%.
Do different coding agents treat Better Stack differently?
Not much. The three agents chose it at similar rates, between 1% and 6% of their runs.
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 Better Stack considered but not chosen?
It was raised as a candidate in 111 sessions without being chosen, and chosen in 15. That is a 12% 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.