How to get picked for databases by coding agents
Neon took 46% of 732 judged databases sessions. What the numbers say a vendor in this category should do.
If you sell databases, this page is the part of the market no dashboard shows you: what a coding agent does when a developer asks for databases and never compares vendors.
The numbers come from 732 judged sessions with Claude Code, Codex and Cursor, spread across 24 realistic codebases, with every session read by a judge.
What coding agents choose for databases
Across 732 judged sessions, Neon was chosen most often, in 46% of runs. Supabase was second with 37%.
| # | Product | Runs won | Share |
|---|---|---|---|
| 1 | Neon | 334 | 46% |
| 2 | Supabase | 269 | 37% |
| 3 | Turso | 48 | 7% |
| 4 | Azure Database for PostgreSQL Flexible Server | 21 | 3% |
| 5 | Firebase | 17 | 2% |
| 6 | Google Sheets | 10 | 1% |
| 7 | Netlify Forms | 6 | 1% |
| 8 | Formspree | 4 | 1% |
| 9 | Render Postgres | 4 | 1% |
| 10 | Azure SQL Database | 3 | 0% |
Full board, every run replayable: the databases leaderboard.
What the shape of this category means
The leader takes 46% of runs and there is a real second place. The category has a default but it is not settled.
Neon at 46% with Supabase at 37% is a default with a real challenger behind it. The agent is choosing rather than reaching, which means the inputs it uses can move the answer.
The work is to be the easiest correct answer: a quickstart that runs when pasted, documentation that states the current version, and pages that answer the exact configuration questions agents search for.
The agents do not agree with each other
In this category Claude Code, Codex and Cursor did not put the same product first.
| Agent | Runs | Picked most often |
|---|---|---|
| Claude Code | 380 | Neon (184) |
| Codex | 181 | Supabase (95) |
| Cursor | 171 | Neon (82) |
That split decides where a vendor spends. In this category Codex ran a web search in 96% of its sessions and Claude Code in 0%, so what you publish reaches nearly all of Codex's databases sessions and almost none of Claude Code's. Taking Supabase's position with Codex is mostly a content problem. Taking Neon's with Claude Code leans more on the repository.
Who is asking changes the answer
Every request in this experiment was written as a specific kind of person. In this category the leader changes with the person.
| Who is asking | Runs | Picked most often |
|---|---|---|
| Vibe coder | 417 | Supabase |
| Junior developer | 163 | Neon |
| Senior engineer | 113 | Neon |
| Enterprise team | 24 | Azure Database for PostgreSQL Flexible Server |
That is 3 different products winning databases for 4 kinds of buyer, out of the same 732 sessions. Nobody here is winning databases. They are each winning one kind of buyer.
If you sell to more than one of them, you need pages for each. See how to win the enterprise persona.
What you are really competing against
In this category agents never chose to build it themselves. All but 2 sessions ended with a product. That is good news: you are in a straight vendor comparison, and the levers that work are the ones you control.
Considered, and never chosen
Because the judge records every product an agent raised and not only the one it picked, this board also shows who kept reaching the shortlist and losing. In databases the clearest case is PGlite: on the table in 255 sessions, chosen in none.
| Product | Raised in | Chosen in |
|---|---|---|
| PGlite | 255 sessions | 0 |
| PlanetScale | 100 sessions | 0 |
| Convex | 88 sessions | 0 |
| MongoDB Atlas | 55 sessions | 0 |
| expo-sqlite | 42 sessions | 0 |
Being rejected is a better position than being unknown, and a cheaper one to fix. The product is already in the agent's head and on the list. Whatever ended those 540 sessions is recorded in each transcript, one reason at a time.
What to do about it in databases
- Skip the build-versus-buy argument. No databases session in this experiment ended with the agent writing its own implementation. All but 2 adopted a product, so the whole contest is against the other names in the table above.
- Aim at second place first. Neon holds 46% and Supabase holds 37%. The gap between the default and the field is where the reachable sessions are.
- Pick which buyer you are for. The same databases need written as a vibe coder landed on Supabase, and written as an enterprise team landed on Azure Database for PostgreSQL Flexible Server. Those are two markets, and the enterprise one needs pages containing the constraint words: audit log, data residency, retention, single sign-on. See how to win the enterprise persona.
- Measure per agent. Claude Code put Neon first, Codex put Supabase first, Cursor put Neon first. A blended number for databases describes a market that does not exist.
The work that applies to every category rather than to this one is written up separately: audit your documentation, write a quickstart an agent can follow, and how to measure install share.
Every database on this board
One page per product, with its install share, the per-agent split, and how often it was raised without being chosen.
- Do coding agents recommend Neon? — chosen in 46% of sessions
- Do coding agents recommend Supabase? — chosen in 37% of sessions
- Do coding agents recommend Turso? — chosen in 7% of sessions
- Do coding agents recommend Azure Database for PostgreSQL Flexible Server? — chosen in 3% of sessions
- Do coding agents recommend Firebase? — chosen in 2% of sessions
- Do coding agents recommend Google Sheets? — chosen in 1% of sessions
- Do coding agents recommend Amazon S3? — chosen in 0% of sessions
- Do coding agents recommend SQLite? — raised in 524 sessions, chosen only within 1 joint choice
- Do coding agents recommend PGlite? — raised in 255 sessions, chosen in none
- Do coding agents recommend PlanetScale? — raised in 100 sessions, chosen in none
- Do coding agents recommend Convex? — raised in 88 sessions, chosen in none
Where these numbers come from
732 judged sessions in databases across 24 codebases, part of a published set of 15,287. Real coding agents at pinned versions, in sandboxes, inside realistic codebases, with a simulated project owner in the loop and a blind judge on every session. The full method is on one page: how we measured this.
Every databases run can be replayed on the board.
<!-- generated by scripts/write-data-pages.mjs -->
Common questions
How many codebases is this based on?
732 judged sessions across 24 realistic codebases. A category only runs on repositories where its seam is open, so coverage differs: some categories ran on more than ten codebases and some on two.
What database do coding agents choose?
Across 732 judged sessions, Neon was chosen most often, in 46% of runs. Supabase was second with 37%. The result changes by agent and by who is asking.
Do Claude Code and Codex pick the same database?
No. Claude Code picked Neon, Codex picked Supabase, Cursor picked Neon. Measuring one agent tells you about part of the market only.
How often do agents build databases themselves instead of installing something?
Never, in this category. No session ended with the agent writing the code itself. All but 2 adopted a product; the rest used built-in tools or made no choice.
How can a vendor improve its position here?
Make the quickstart run when pasted, state the current version on the documentation page, use one name across product, package and import, write pages for the symptoms users describe rather than only the category name, and get into the repository through templates and framework integrations.
Which databases do agents consider but never choose?
PGlite (raised in 255 sessions, chosen in none), PlanetScale (raised in 100 sessions, chosen in none), Convex (raised in 88 sessions, chosen in none), MongoDB Atlas (raised in 55 sessions, chosen in none), expo-sqlite (raised in 42 sessions, chosen in none). Being considered and not chosen is a different problem from being unknown, and it is usually fixable.
Where this comes from
Armature ran 15,287 judged sessions with Claude Code, Codex, Cursor, Grok Build CLI and Muse Code inside 110 realistic codebases, and published every run. The numbers on this page come from that work.