customer_support · finance · workflow
NewDay builds a generative AI-based customer service agent assist with over 90% accuracy
NewDay's contact center handles 2.5 million calls annually, and with nearly 200 knowledge articles agents had to manually search during live customer calls, slowing resolution times for both agents and customers.
How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · Agent submits question via UI
A customer service agent logs in and submits a question to the NewAssist UI.
Outcome
NewAssist achieved over 90% accuracy for agent question answering and reduced answer retrieval time from an average of 90 seconds to 4 seconds; the solution is currently rolling out to over 150 agents across Customer Operations.
What failed first
An initial full voice-assistant concept was abandoned as too ambitious, and an early text-parsing approach using PyPDF performed at only around 60% accuracy because it did not account for the widget-based structure of NewDay's knowledge articles.
Results
Time saved90 seconds
Volumeover 90%
Cost replacedunder $400 per month
Grounding & classification
Source type: technical build writeup
31 fields verified against source quotes, 13 dropped as unverifiable.
agent assistknowledge searchragknowledge basehuman review describedmetric backednamed customerproduction runtime claimedvendor confirmedworkflow describedfinancial servicesaccuracy improvementcost reductionemployee productivityresolution time reductiontechnical build writeupcall center aicustomer supporthuman review queuerag answering