Event-Based Customer and Business Intelligence
Call Center Analytics
Our ground-up analytics work drove a call center's answer rate from 26% up to the theoretically maximum possible of 65%, increasing NPS scores as well.
The challenge
A large call center client was experiencing a serious problem. They began to receive feedback and negative NPS scores with a recurring theme: callers were not able to reach live agents. For a white-glove service provider whose primary value proposition is access to skilled human agents, that gap between promise and experience was critical.
The first problem was that they didn't have any quantitative evidence to back it up. There was absolutely no data around incoming calls, call answer rate, or anything of that nature — only anecdotal complaints and slipping satisfaction scores.
Building the data foundation
We built a data extraction, transformation, and load process, along with a full data warehouse, to pull all call data in from the client's WebEx voice-over-IP call system.
We married that data with their internal team-based analytics to create views by manager, team, and department — giving leadership the first clear picture of how answer performance varied across the organization.
What the numbers revealed
After connecting call data to team structure, the reality was stark. Cold answer rates on some teams were as low as 10%, and the overall average sat at 26%. With a theoretical maximum based on agent availability of 65%, there was a long way to go — but now there was a baseline everyone could see and act on.
We created the first version of analytics surfacing what actual call rates were by department, team, manager, and individual call center representative.
From visibility to results
The CEO mentioned the new tracking at the annual company kickoff. Just by making the data visible, answer rates climbed about 15% within the first week.
We operationalized these analytics with daily check-ins, then weekly check-ins. With constant work over about a year, the organization reached the theoretical maximum of 65%. Negative feedback about calls going unanswered all but went away, and NPS scores went back up to their normal levels.
ETL pipeline and data warehouse for WebEx call data
Answer rate views by department, team, manager, and representative
Operational cadence with daily and weekly performance check-ins
Answer rate improvement from ~25% to the 65% theoretical maximum
Restored NPS scores and elimination of unanswered-call complaints