Hey everyone! 👋 I've been lurking here for a bit, but this is my first real post. I'm a data analyst on a team that recently migrated our internal analytics Slack bot from a legacy custom tool to Claw. We were super excited about the new features, especially the natural language querying.
The migration itself went *mostly* smooth for historical data, but we're hitting a weird snag. Post-migration, when we ask the new Claw bot questions about recent projects (like, "What was last week's conversion rate for Project X?"), it often comes back with "I don't have enough context" or pulls data from *before* the migration cutover date, ignoring the last 7-10 days. It's like it can't see the most current data in our data warehouse, even though our Looker models are connected and updated.
Has anyone else run into this? I'm trying to figure out if it's:
* A configuration issue on our end with the data source connection in Claw.
* Something about how Claw's bot caches or indexes data post-connection.
* A common hiccup during the cutover phase that we just need to wait out.
We followed Claw's migration docs, but maybe we missed a step for syncing the "context" or recent history. Any tips or walkthroughs from teams who've been through this would be amazing! My team is leaning on me to fix this, and I'm still getting my bearings with the new tool.
That's a common symptom of a stale data index, not a broken connection. Claw's bot builds an internal search index from your Looker models on a schedule, separate from the live data connection. After a migration, it often defaults to a weekly refresh.
Check the "Context Indexing" tab in your Claw workspace settings. You'll likely see the last successful index job was right before your cutover. Run a manual refresh for your project's models. It can take a few hours to process new data.
If that doesn't clear it, their support usually points to a specific date filter being set in the model's exploration permissions that the bot inherits, which can exclude recent days.
—AF