We are aware of an ongoing issue affecting LLM-powered analysis features — brand research, brand info, topics, queries, and competitor analysis — that began around May 29. Analyses may appear stuck in a pending or loading state and will not return results; no action is needed on your end.

## 🔍 What's Happening

Since approximately May 29, the LLM analysis pipeline that powers several Search Atlas features has been interrupted. Affected features display results as permanently pending or loading with no output delivered.

The following features are currently affected:

- Brand Research
- Brand Info
- Topics
- Queries
- Competitor Analysis

If you submitted any of these analyses on or after May 29, your job is queued and will be processed automatically once service is restored. You do not need to re-submit.

## 🛠️ Workaround Steps

There is currently no workaround available for this issue. Do not re-submit your analyses — resubmitting will not resolve the problem and may create duplicate entries in the processing queue.

All queued analyses will run automatically when the pipeline is restored.

## 📡 Engineering Status

Our engineering team is investigating the root cause and is actively working to restore the pipeline. All analyses submitted since May 29 will be processed once service resumes — no data or requests will be lost.

This article will be updated as soon as the fix is deployed and results begin delivering to affected accounts. We will provide an estimated timeline as soon as we have more information.

## 📞 When to Contact Support

Reach out to our support team via the in-app chat if any of the following apply:

- You have a time-sensitive deadline that cannot wait for the pipeline to restore.
- You see no resolution or status update after 24 hours.
- You are experiencing a separate issue not described on this page.

When you contact support, include your account name and the date you submitted the affected analysis so the team can assist you faster.

**🌀 Our team is fully engaged on restoring the pipeline and will process all queued analyses as soon as service resumes — we sincerely apologize for the disruption and appreciate your patience.**