What Our Call Centers Want Us to Fix

I took on a new team in late Spring, 2025. They hadn’t had a manager for about a year because the team was very good at making tactical changes to keep up with 'business as usual'. In some ways, this is a compliment, but my observation was entirely different - this team had a lot of untapped potential. They could become partners to our product team, not just deliver wireframes on demand. 

Occasionally, you have to place a few bets to energize your best workers, to give them something new or out of the ordinary to help them fall in love with design again.

Based on past success, I asked this new team to help me really understand the voice of the customer, not just from complaint data or completion rates. What would a customer tell me if she was sitting next to me. 

A talented designer worked with me and a talented researcher and devised an audacious plan. We’d embed in call centers to hear first hand what people were calling about in their own words. The goal was to go from what was happening, to why it happened. We had the data to prove the value - about 67% of calls into the call centers were for an issue in my new domain.

Over the course of a week and a half, we visited call centers across the United States, each focused on one product in our portfolio. 

Each of the three of us would listen to calls for an hour. I liked to start by asking the agent to predict what calls they were expecting. After sitting for an hour, we’d move to another specialist in the same unit. After two hours, and perhaps 15 calls, we then pulled up, the design team and the 6 agents we had just observed. We asked them to help us to separate the one-offs from the true customer problems. And they did that with burning desire to help.

They helped us write customer journeys, which have proven to be very useful. Our approach was to give them a lot of clip art that they could cut out and assemble, an example below.

They were motivated because they saw us as their mouthpiece, their champions. At the end of the sessions, we had listened to some 500 calls and spoke to about 50 agents. We knew we had the right pain points, because we asked our specialists repeatedly, and they kept confirming what we said.

When we got home, it took some time to synthesize, but we came up with some great themes. For example, about sixty percent of the calls into one call center are “misroutes,” meaning they could not be resolved by agent. Why? Likely because of a reorg and our UI had not kept up, but if a call costs $2.53 (which it does), then that’s a lot of savings in one day.

Here’s one of a handful of slides we shared with our product executive:

But there were many more, some easy to fix, others harder. Some were general, some product specific. In the end, there were 16 themes. We shared them with product executives one at a time. At some point, it became clear that we had moved the needle. That day came when our business strategy team wanted to help us put dollar signs next to the money Chase spends per year on these issues.

Below is the money shot. It showcases the cost/year of these pain points. We have also aligned these to JD Power issues and t-shirt sizing for tech lift. The literal dollar amounts are blacked out, but this slide and two more just like it are making the rounds on how my team set a new bar for design (in a team of 1200 designers) and we have been asked to repeat this process with other teams.

The spend we have identified is in the millions of dollars / year, and our product executive is working to get incremental teams to address.