Redesigning how customers start, stop, and transfer utility service

Overview

A digital option already existed. Most customers still picked up the phone.

Background

Avangrid serves millions of utility customers across the Northeast. The contact center was the most used channel for starting service, and the one with the lowest satisfaction.

Problem

80% of customers called to start, stop, or transfer their service, despite an online option. 30% needed more than one contact to activate service, and 44% spent more than ten minutes setting up and confirming a start date.

My role

I led foundational research with current customers, mapped what happened inside Avangrid after each request, ran a cross-functional strategy workshop, then designed and tested two competing concepts before setting the final direction. I worked with Avangrid’s CX, product, operations, and engineering teams.

Results

• 80%Shift to digital
  The redesigned journey became part of Avangrid’s broader CX transformation, which shifted 80% of customer interactions to digital channels.
• 200%NPS improvement
  The move-in / move-out redesign contributed to a 200% increase in NPS across customer journey channels.
• 2024E Source award
  The work set the framework and patterns later extended to outages, billing, and account, and the program went on to win an E Source award.

Challenge

Make the digital option the one customers trust, even in the middle of a move.

A routine task in the most expensive channel

Most start, stop, and transfer requests ran through the contact center, the most expensive channel and the lowest rated. Three in ten customers didn’t even know signing up online was an option, and more than half said they had never visited the website.

A website that still ran on manual work

Online requests landed in an email queue and were re-keyed into back-office systems by hand. Identity checks were manual, address data wasn’t synced with USPS records, and online account creation lagged behind the request. Customers felt the delays even when they couldn’t see the cause.

No receipt, no confidence

Customers treated a confirmation email like a receipt. When it was delayed or missing, anxiety rose quickly and the phone felt like the safer bet, even for customers who preferred digital tools. Moving made it worse. They were coordinating dates, addresses, landlords, bills, and other services all at once, so small gaps, like online start dates blocked three days out and on weekends, felt much larger.

Three journeys, multiple companies

The move in, move out, and transfer service design solutions needed to be flexible enough to extend to each of Avangrid’s child companies.

Approach

This was a service design problem: research the customer touchpoints and what’s behind the curtain, reframe the journey, then test competing ideas.

Understand

Learn why customers called, and what happened behind each request. I conducted 27 customer interviews (more than 1,600 minutes), gathered existing survey and analytics data, conducted stakeholder interviews, and created process maps from internal documentation and training materials.

Mapping the system behind the customer journey

Reframe

Redesign the journey before, during, and after the transaction. I brought customer insights, operational findings, and cross-functional partners together in a strategy workshop, then prioritized recommendations by value and effort with Avangrid stakeholders.

Designing for confidence

Validate

Design and test two divergent directions before converging. I led sessions with eight customers who walked through both concepts and compared them directly. The final direction paired the clearer flow of one with the reassurance features of the other.

Two concepts, one stronger final direction

Key moments

Three moments that shaped the final design.

Mapping the system behind the customer journey

We mapped six current-state journeys (start, stop, and transfer, across both the website and the call center) in layers: customer actions, internal activities and systems, pain points, and opportunities.

The maps exposed problems customers could feel but not see. Web requests still needed manual processing. Identity verification depended on manual steps. Address data wasn’t fully synced with USPS records, confirmations were inconsistent, and service blocks appeared without an explanation or a clear path forward.

A line from Avangrid leadership in our strategy workshop became a guiding principle for the rest of the project:

“I don’t want to create a pretty interface that has messed up processes behind it.”

Designing for confidence

Customers didn’t just want convenience. They wanted confidence. Channel adoption first seemed like a usability problem, but interviews with customers who recently started, transferred, or stopped service told a more nuanced story. The phone confirmation felt like a receipt: talking to a person gave customers confidence that their request was received and understood. Even customers who preferred digital had been let down before, so a late or missing confirmation sent them back to the phone. Reassurance mattered as much as speed, and the stress of moving can amplify small uncertainties.

Three behavioral archetypes emerged from the research. They helped the team design for different attitudes instead of forcing everyone into one channel, and were circulated across the organization to build awareness and customer empathy.

To address these archetypes, we set several design principles:

• Design reassurance into every step.
• Set expectations before asking for effort: what customers need and how long it will take.
• Confirm immediately, then make status and next steps easy to find later.
• Keep support available for customers who need it.

Two concepts, one stronger final direction

Rather than refine one solution too early, we designed two distinct concepts for starting service. One was a linear flow with a single question per step. The other was a hub approach, with grouped questions in collapsible sections and more visible progress.

I organized and conducted one-on-one sessions with eight customers, who walked through both concepts in counterbalanced order and then compared them directly. The goal was to understand why each approach worked so we could combine the strongest elements.

Customers preferred the linear, one-question-at-a-time approach but liked several specific pieces of the hub approach. Testing showed the design was on the right track:

• No major blockers
• Expected under 10 minutes
• Clear expectations reassured
• Time estimates built confidence
• Photo ID verification delighted
• Bill estimates struck a chord

Impact

Trusted self-service channels and a template for future change.

We delivered validated responsive designs for move in, move out, and transferring service, including complex steps like identity verification via capturing a photo of a driver’s license or passport. The work shipped, and the same methodology and approach carried into the subsequent journey designs, which I led.

80% of customer interactions shifted to digital channels across the broader CX transformation.