Quality assurance · Production

PetDesk raises Guru knowledge verification score from 76% to 85% with quality automation

The problem

After years of growth through acquisitions, PetDesk's Guru workspace held duplicate and conflicting documentation across merged teams, and the unverified queue was so cluttered with timed-out cards that experts stretched thin could not identify the content that actually needed updating.

Workflow diagram · grounded in source
1
Default quality rules activated
trigger
“Shona turned on Guru's default quality rules and gave it a day. Within 24 hours, it was immediately apparent which rules needed adjusting to suit their org”
2
Cards auto-verified by quality rules
validation
“the reasons the automations unverified or verified their sources”
3
Custom thresholds calibrated for org
validation
“flagging answers with more than two thumbs-down in the past month gave Shona a clear, undeniable signal that something in that card needed attention — no debate required. Second, excluding admin views from the "viewed frequently" thresho…”
4
Real problems surfaced in queue
output
“what's unverified now - it's not just noise anymore — it's the stuff that actually needs love”
5
Collection owners review flagged cards
human_review
“When PetDesk's Director of Product saw a focused list of what needed attention — not a sprawling review queue — she enthusiastically reorganized her entire collection, merged legacy cards, and assigned clearer ownership going forward”
6
Knowledge improved for AI agents
output
“adding brief descriptions to the top of Cards to help Knowledge Agents easily discern the purpose of Cards”
7
Automation expanded to more agents
feedback_loop
“Quality automation expanded to two Knowledge Agents — with additional agents being piloted across the org”
Reported outcome

Guru quality automation raised PetDesk's verification score from 76% to 85%, cleared the unverified queue of noise so it reflects real gaps, consolidated eight duplicate risk management cards into one cohesive process, and prompted stakeholders who had not engaged with Guru in years to take ownership of their content.

Reported metrics
Verification score76% to 85%
Historical verification score range (pre-automation)60-70% range for years
Duplicate risk management cards consolidatedeight consolidated into one
Knowledge Agents with quality automationtwo
Reported stack
GuruKnowledge Agent
Source
https://www.getguru.com/customers/petdesk
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Guru quality automation raised PetDesk's verification score from 76% to 85%, cleared the unverified queue of noise so it reflects real gaps, consolidated eight duplicate risk management cards into one cohesive process…

What tools did this team use?

Guru, Knowledge Agent.

What results were reported?

Verification score: 76% to 85%; Historical verification score range (pre-automation): 60-70% range for years; Duplicate risk management cards consolidated: eight consolidated into one; Knowledge Agents with quality automation: two (source-reported, not independently verified).

How is this quality assurance AI workflow structured?

Default quality rules activated → Cards auto-verified by quality rules → Custom thresholds calibrated for org → Real problems surfaced in queue → Collection owners review flagged cards → Knowledge improved for AI agents → Automation expanded to more agents.

Related quality assurance cases