Call center ai · Production

Thryv turns directory assistance into a 24/7 channel with Voice AI via Synthflow

The problem

Thryv's directory assistance service was staffed by human agents limited to standard business hours, creating a structural gap where after-hours demand went unserved. Extending coverage to 24/7 was not economically viable with a traditional staffing model, resulting in missed calls and untapped demand.

Workflow diagram · grounded in source
1
Caller dials directory assistance
trigger
“a New Zealand-based directory assistance line that helps callers find residential and business numbers”
2
Voice AI retrieves directory info
ai_action
“The agent was designed to retrieve and deliver information in real time via API integrations”
3
Edge-case escalation to human
routing
“complex or edge-case requests could seamlessly escalate to human agents when needed. This ensured reliability without sacrificing coverage.”
4
Call answered and served
output
“All inbound calls are now answered, including after-hours and periods of concurrent demand — eliminating missed calls that were previously unavoidable under a human-only model”
Reported outcome

Thryv's Voice AI agent now handles over 20,000 directory assistance calls per month at a 100% answer rate, recovering 4.4% more calls that previously went unanswered, and reached live production in approximately one month.

Reported metrics
Directory assistance calls handled per month20K+
Call answer rate100%
Additional calls answered+4.4%
Time to production~1 month
Reported stack
Synthflow
Source
https://synthflow.ai/customer-stories/thryv-turns-directory-assistance-into-a-24-7-channel-with-voice-ai
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Thryv's Voice AI agent now handles over 20,000 directory assistance calls per month at a 100% answer rate, recovering 4.4% more calls that previously went unanswered, and reached live production in approximately one m…

What tools did this team use?

Synthflow.

What results were reported?

Directory assistance calls handled per month: 20K+; Call answer rate: 100%; Additional calls answered: +4.4%; Time to production: ~1 month (source-reported, not independently verified).

How is this call center ai AI workflow structured?

Caller dials directory assistance → Voice AI retrieves directory info → Edge-case escalation to human → Call answered and served.

Related call center ai cases