Teammate review monitor: Remove convo from review list, change reviewee | Community
Skip to main content
Submitted

Teammate review monitor: Remove convo from review list, change reviewee

Related products:Fin & AI
  • Oct 20, 2026 1:01 PM UTC
  • 3 replies
  • 23 views

Hiya!
 

We have 2 feature requests for the "Teammate reviews" monitors:

  • A few times, we've run into a situation where we'd like to remove a conversation from the review, either because nothing happened in the convo (e.g. customer just dropped off without saying anything), or because the conversation was so trivial as to not be worth reviewing at all.
  • We have also occasionally wished to change who the reviewee is on a review -> it looks like it assigns the reviewee based on who closed the conversation, but we do have cases where one person takes the majority of a conversation, and then for whatever reason, it's reassigned to someone else who does the last little bit and then closes it. It would be great to have the option to reassign from the review panel.


Third bonus request:

  • Having to click update on every single manual criteria is a bit annoying, especially on criteria where there are multiple ratings reasons -> if it's not possible to remove the requirement to click "update", then at least it could be moved higher up on the list, or the ratings reasons could be "foldable" or something. For me with a huge monitor it's no problem, but for other colleagues working with smaller screens it's frustrating enough to discourage regular use :)

3 replies

Forum|alt.badge.img+3
  • Active User
  • Oct 26, 2026 8:04 AM UTC

These requests make sense, especially removing conversations that don't need review and allowing the reviewer to be reassigned.

 

I'm curious: how does your team decide which conversations should be reviewed in the first place? Is it mainly random sampling and manual selection, customer escalations, or specific signals such as incorrect promises and privacy concerns?

 

I'd be interested to learn how teams identify the conversations that genuinely need attention without having to review everything manually.


  • Author
  • New Participant
  • Oct 27, 2026 3:14 PM UTC

Hi Parvesh! Thanks. 

 

I have it set up currently to do automated QA reviews of general teammate performance, randomly sampled across the team inboxes they’re active in - not specifically QA for flagged cases of escalation, incorrect answers, or privacy issues. The reason for doing it automatically & randomly is to avoid sampling bias by the person running the QA.

AFAIK there’s also no way to manually select a conversation to be added to a team QA monitor.


Forum|alt.badge.img+3
  • Active User
  • Oct 27, 2026 3:43 PM UTC

Thanks for explaining! That makes sense. Random sampling is useful for avoiding bias when evaluating overall teammate performance.

 

It sounds like there's a distinction between general QA sampling and identifying specific conversations that may need attention because of an incorrect promise, a policy violation, or a privacy concern.

 

I'm building ChatShield to flag potentially risky conversations involving both human agents and AI for human review, so it could complement general QA rather than replace it.

 

When your team discovers a conversation involving one of these specific risks, how do you currently make sure it's identified and reviewed? Is that handled through a separate process, or does it depend on whether the conversation happens to be selected in the random sample?