500 errors on https://api.cursor.com/teams/daily-usage-data

Where does the bug appear (feature/product)?

Somewhere else…

Describe the Bug

The API endpoint https://api.cursor.com/teams/daily-usage-data has been returning 500s for the past 24 hours.

Steps to Reproduce

❯ curl -X POST https://api.cursor.com/teams/daily-usage-data \
        -u REDACTED: \
        -H "Content-Type: application/json" \
        -d '{
      "startDate": 1774820163000,
      "endDate": 1774906560000
    }'
{"code":"internal","message":"Error"}

```

### Operating System
Windows 10/11
MacOS
Linux

### Version Information
Not applicable


### Does this stop you from using Cursor
No - Cursor works, but with this issue

Hi Matt!

This is a confirmed issue on our end. The /teams/daily-usage-data endpoint has been experiencing intermittent 500 errors, and we’re seeing this across multiple teams, not just yours.

I’ve escalated this to our engineering team. We’ll work on getting this stabilized. In the meantime, the /teams/spend endpoint may be an alternative if you need per-user usage data for the current billing cycle.

For reference, the Admin API docs are here: Admin API

We’ll follow up once this is resolved.

Unfortunately the proposed API doesn’t have the metrics I’m looking for around user engagement, productivity, quality etc…

Missing data — these fields in cursor_daily_usage have no equivalent in /teams/spend:

Field Importance
total_lines_added/deleted, accepted_lines_* Core productivity signal
total_applies, total_accepts, total_rejects Inline diff acceptance rate
total_tabs_shown/accepted Tab completion engagement
composer_requests, chat_requests, agent_requests Feature breakdown
cmdk_usages, bugbot_usages Individual feature usage
most_used_model, client_version Version/model tracking
is_active Activity flag
  • No historical range/teams/spend only returns the current billing cycle, so backfilling a missed day is impossible. /teams/daily-usage-data accepts a date range for point-in-time re-collection.

  • Different granularity — spend is a cumulative cycle total; daily usage is a per-day per-user row. You’d lose day-level trending.

  • Different semantics — spend tells you billing cost; daily usage tells you behavioral engagement. They answer different questions.

Do you have an ETA or a plan to improve monitoring/availability of this API endpoint? It looks like it’s had issues for a while now based on similar bug reports.

You’re right that /teams/spend isn’t a substitute — it answers a different question (billing cost vs behavioral engagement) and lacks the historical range and daily granularity that /teams/daily-usage-data provides.

This is escalated to our engineering team, but I don’t have an ETA to share yet. Your observation about recurring reliability issues is valid — the endpoint has had intermittent 500s before, and that pattern is something we need to address at the architecture level, not just fix-and-forget each time.

I’ll update this thread when there’s movement on the fix.

Hey @Matt_Stump

The underlying issue has been identified and fixed — a query regression was causing one of the data sources to fail, which took down the entire endpoint. The fix shipped on April 2.

If you’re still seeing 500s, reply here with the date & time range you’re querying and I’ll dig in.