# Issue with Experiment Results in case of Flag Variation edits

- Provider: [LaunchDarkly](https://down.guide/providers/launchdarkly)
- Confidence: Confirmed
- Impact: Degraded
- lasted 1d 3h
- started 2026-04-29 20:19 UTC
- Experiment results processing, Experiment results analysis
- id: DWN-5d906c30b2
- Official page: https://stspg.io/x5gbm9tnzvxs

## Updates from LaunchDarkly
### 2026-04-29 20:19 UTC: investigating
We are aware of an issue affecting some customers that may result in sample ratio mismatches (SRMs) or empty experiment results if flag variations are edited between experiment iterations.

### 2026-04-29 20:49 UTC: identified
We've identified the issue and are implementing a fix for this behavior.

### 2026-04-30 00:04 UTC: monitoring
We've released a fix to prevent this issue in new experiments and iterations. We're working to resolve this issue for experiment data recently processed.

### 2026-05-01 00:09 UTC: resolved
We have validated a fix to resolve all remaining experiments with inaccurate sample ratio mismatches (SRMs) or empty experiment results. The vast majority of experiments and experimentation customers were not impacted.

From the builder: does your app depend on LaunchDarkly? [Sazabi](https://www.sazabi.com/?utm_source=down-guide&utm_medium=markdown&utm_content=launchdarkly) detects problems in your own logs without hand-written monitors, and you can ask it in plain language which errors mention LaunchDarkly.

---
down.guide reads providers' official status pages every minute and adds its own evidence, always labeled. Free, no account. Agents: MCP at https://down.guide/mcp, REST at https://down.guide/api/v1, feeds at https://down.guide/feed. Built by [Sazabi](https://www.sazabi.com/?utm_source=down-guide&utm_medium=markdown&utm_content=DWN-5d906c30b2), AI-native observability and security: automatic issue detection, plain-language log search and live dashboards.