PURA hearing probes Eversource customer-service metrics, Stella Connect reporting and contractor pay-for-performance
Get AI-powered insights, summaries, and transcripts
Sign Up FreeSummary
Eversource's customer panel told PURA it provided Stella Connect monthly scores but not a single final narrative report; the panel also detailed a five-tier pay-for-performance (PBR) contract with Alorica and discussed call-quality metrics, silent time and how missed-appointment fees are handled.
Connecticut’s utility regulator pressed Eversource witnesses on Oct. 12 about two interrelated customer-service issues: 1) the company’s reporting from the Stella Connect customer-feedback tool after the tool’s pilot ended, and 2) the new performance-based contract with call-center vendor Alorica that ties a portion of vendor pay to quality metrics.
Stella Connect: Eversource told PURA that its Stella Connect engagement produced per-employee, per-interaction star ratings and comment text but did not generate a single summarizing ‘‘final report’’ of the sort PURA staff had requested. Jared Lawrence, senior vice president and chief customer officer, and Tim McGrath, director of contact centers, explained they receive aggregated monthly and annual scores that are what leadership reviewed and that the vendor did not supply a bundled narrative or capstone report. The company said it can produce monthly rolls and customer comments on request; staff pressed the company to provide the data it had been using for leadership reporting.
Alorica pay-for-performance contract: Witnesses described a PBR structure under which Alorica earns based on tiers tied to measured performance. Metrics in the contract include hold time (the average time a customer is put on hold after a CSR engages), silent time (dead air on a call), call-quality scores (graded on several criteria), consultant-resource hours delivered, and other operational measures. Eversource and the vendor perform frequent calibration sessions (three per week, witnesses said) to align scoring. The company said call-quality sampling uses at least four evaluated calls per CSR per month and that the PBR aggregates those results to produce a single quality component that contributes to monthly vendor pay.
Escalations, complaints and missed-appointment fees: The company’s complaint-handling process includes regulatory-specialist review and the ability to listen to recorded calls when a consumer alleges a problem. For certain events, the company said staff will reclassify a complaint as ‘‘justified’’ and take remedial steps; Eversource said one of its human-performance processes flagged instances where it could not substantiate whether a customer had contacted the company prior to being charged an inspection/missed-appointment fee (the commonly referenced $90 charge). The company said it has added coordination between contractors and company staff to reduce assessable missed-appointment fees.
Why this matters: These operational practices influence customer experience and can affect enforcement outcomes, compensatory remedies, and the regulator’s assessment of whether the utility meets its service-quality obligations.
What parties asked for: PURA and intervenors asked for supplemental materials that the company said it could produce — monthly Stella scores/comments and additional detail about the PBR metrics and calibration. The company agreed to provide a written technical supplement on how shared reference tables in Eversource’s customer information system create cross-operating-company dependencies and to file certain Excel read-ins after consulting development teams.
Key quotes: "Stella was a tool...it does not generate routine reports. So we used it to provide a weekly and monthly average figure of how our CSRs were performing," Jared Lawrence said. "If there are performance issues, they're then reviewed with the CSR as part of coaching," an Alorica contract manager said.
Operational follow-ups: PURA asked the company to provide additional detail about the missing Stella-style ‘‘final report’’ and to produce supporting data for certain sample calls and complaint-resolution examples.
