cancel
Showing results for 
Search instead for 
Did you mean: 
Genie Hub
Explore technical articles, practical guides, best practices, and real-world use cases to help you get the most out of Databricks Genie. Learn from the Databricks team, MVPs, and community experts.
cancel
Showing results for 
Search instead for 
Did you mean: 

Trust in Genie isn't won with accuracy — it's won with a visible feedback loop accountability

Valeria_Koz_DBX
Databricks Employee
Databricks Employee

A Finance team won't trust a Genie Room just because it's right. They trust it when they see someone is accountable for the answers.

What worked for us:

  1. a thumbs up/down on every chat, and total transparency about who reads it. We had a dedicated Finance team owning the rollout — they knew the real questions and investigated every miss.

  2. and a thumbs-down doesn't just sit in a dashboard; it goes to the product team, and Databricks genuinely comes back to you on it.

That's the shift: trust isn't won with accuracy stats — it's won when people see a loop. My feedback → a real person → a real change.

How do you close the feedback loop on your Genie rooms — and who owns the answers? 👇

#Databricks #Genie #ChangeManagement

2 REPLIES 2

ivanvyd
New Contributor III

Thanks for sharing, @Valeria_Koz_DBX!

For ownership, I (try to 🙂) favour a named business owner who validates the metric definition and expected answer, paired with a technical owner responsible for investigating the issue and checking the fix.

Confirmed misses can then become self-contained Genie benchmark questions, with reviewed reference SQL where appropriate. Rerunning those after changes gives the team a way to check whether the same issue has returned.

For me, the loop closes with the person who raised it: "Here's what changed and how we checked it. Does this now answer your question?" That gives them a chance to confirm the resolution, rather than just receiving a "fixed" notification.

Appreciate you raising such an important topic.

Ivan Vydrin
Lead Software & AI Engineer · Tech Fabric LLC

Valeria_Koz_DBX
Databricks Employee
Databricks Employee

Hi @ivanvyd 

Completely agree on the two-owner model. For me the accountability piece is the key — putting a name against a metric puts a face to the issue, and suddenly it stops feeling like "the tool got it wrong" and starts feeling human. People trust it more when they know someone actually owns it.

 And your point on closing the fix — agree even more. Being able to cross something off, to say "this is sorted now," is genuinely satisfying on both sides: the person who fixed it and the person who raised it. That loop closing is half the trust right there.