Lab brings the search quality tools into one place. Use it when you want to understand what Layers did for recent searches, investigate weak or zero-result queries, or test text and image search behavior.Documentation Index
Fetch the complete documentation index at: https://docs.uselayers.com/llms.txt
Use this file to discover all available pages before exploring further.
Prerequisites
- Your store must be connected to Layers.
- Your catalog should be synced so test searches use current product data.
- You need dashboard access for the store you want to review.
Open Lab
- In the Layers dashboard, open Lab.
- Use Overview to review recent search activity and current engine configuration.
- Use the Lab tool cards to open Audit Log, Experiments, Test Text Search, or Test Image Search.
What you can review
The Overview page shows recent engine activity from the last 7 days:- Expansions - Search terms Layers added to broaden a query.
- Corrections - Query corrections Layers applied before ranking results.
- Redirects - Semantic redirects that sent a shopper to a configured URL.
- Intent modifiers - Filters or boosts Layers applied after detecting shopping intent.
Replay a search
Some Lab rows include a play icon. Click it to open the same query in Test Text Search. Replay is useful when you want to inspect the result set, Query Understanding, score breakdowns, and demo profile behavior for a logged search.Troubleshooting
Lab is empty
Run a few searches in Test Text Search or wait for shopper search traffic. Lab only shows decisions that have been logged.Recent activity looks lower than expected
The activity cards use a 7-day lookback window. Older decisions may still be available in Audit Log, depending on your account’s data retention settings.A replay opens without the expected context
Replay depends on the logged search query and available session context. If the original request did not include session or customer identity data, Lab may show an anonymous visitor and limited context.Next steps
- Audit Log - Inspect individual AI search decisions.
- Experiments - Review zero-result and weak-coverage searches.
- Test text search - Replay and debug text search results.