MVP to 2.0: Redesigning the Inbox to stop missing leads

CallRail 2021| Architecture Redesign

Hebe Zheng (Product Designer), Pia Kendrick (Product Manager), and the Lead Center Engineers.

Summary

Oct - Dec 2021

Context

Lead Center is CallRail’s communication hub for calls, texts, forms, and chats. After launching as an MVP, adoption lagged. The inbox, the core of the experience, was not reliable or meeting users’ expectations. Support tickets piled up about interactions “disappearing,” and customers described the inbox as “buggy.”

Outcome

I redesigned the inbox to remove hidden logic and align it with how users naturally track work. The new model eliminated support tickets about “missing” interactions, improved adoption of Lead Center, and established the architectural foundation for future soft-phone features.

My Role

As the sole product designer on Lead Center, I owned the project end to end. I aligned the team on the root cause, validated the direction through user testing, and led a phased inbox redesign from concept through release.

Problem

When the Inbox worked against users

The inbox was originally built for phone calls. Its three sections, Incoming, Active, and Recent, mapped cleanly to the lifecycle of a call. But when texts, forms, and chats were added, they were forced into the same structure. Behind the scenes, a series of hidden rules and auto-timeouts moved interactions between sections.

Lead Center was meant to become CallRail’s unified communication hub, but adoption stalled after launch. Customers reported interactions “disappearing,” and support tickets continued to rise. As a result, users couldn’t rely on Lead Center to manage leads putting the product’s long-term adoption at risk.

What I did first

Understanding the inbox

Before redesigning anything, I needed to understand how the system actually worked. I mapped every inbox state, section, and logic rule, working closely with engineers to confirm the backend behavior. Then, I ran comparative usability tests with four existing Lead Center users, showing them the current inbox alongside a prototype.

What we learned: users were icon-hunting. They could not reliably tell what needed attention. Three of four wanted a simpler overall experience.

"I am inferring that someone responded to this form submission, but I'm not sure. It's not apparent that someone took care of something."
— James, Real Estate Investor

Design thinking

Inbox as a to-do list

Three constraints shaped this project from the start. The inbox had to handle a high volume of mixed interaction types. Urgent calls had to remain highly visible. And any change to inbox behavior can’t disrupt CallRail’s call-tracking reporting system.

Instead of managing complexity, I focused on removing it. Group interactions by what users need to do, remove hidden rules and automatic movement, and align the experience with a familiar mental model: a to-do list.

The aha moment: timestamp as natural priority

My initial exploration had three sections: New, Active, and Reviewed. I kept running into a structural problem, an incoming call in New and an ongoing call in Active could push each other out of the view.

That led to the breakthrough. I collapsed New and Active into a single section and let timestamps determine priority. The most recent interaction automatically rises to the top, ensuring urgent items remain visible without additional logic.

Removing a section simplified the solution and eliminated the need for any prioritization rules

One inbox, one source of truth

Previously, new interactions landed in different sections depending on interaction type. In the new model, every interaction enters the same place — Active. This created a single, predictable source of truth and removed the ambiguity that made interactions feel “lost.”

Removing auto-timeout and the tradeoff

The most important structural change was eliminating the automatic rules that moved interactions between sections. Now interactions remain in Active until the user marks them Reviewed. This matched the to-do list mental model users already had.

‍Tradeoff:
This approach introduced one risk: Active could become crowded. In usability testing, users didn’t always move items to Reviewed. I advocated for a bulk “Mark as Reviewed” action to reduce friction, but it was scoped out for launch.

Getting it shipped

Org Alignment: The Inbox Roadshow

Because this changed a fundamental product behavior, I partnered with my PM to run an internal alignment tour, the Inbox Roadshow, before anything was officially committed on to the roadmap. We walkthrough the current system logic, user research findings, and proposed structural changes.

This helped teams understand both the technical complexity and the user impact, building alignment across engineering, product, and leadership.

Impact 🎉

After launch, support tickets about “missing” interactions were eliminated.

The simplified inbox architecture also became the foundation for several soft-phone capabilities shipped later that year, including Call Waiting and Queue Callback features that would have been difficult to build on top of the previous system.

The redesign effectively graduated Lead Center from an MVP to a product customers could trust.

Reflection

This project ran unusually smoothly because alignment started early.

By collaborating with engineers while mapping the system logic and presenting proposals as interaction flows rather than polished UI, we avoided late-stage surprises and kept the team focused on solving the right problem. Visualizing the complexity of the existing inbox also helped the broader organization empathize with the user experience.

If I were to revisit this project, I would push harder for a bulk “Mark as Reviewed” action. The need surfaced during usability testing, and post-launch feedback later confirmed it as users reported that the Active section could feel crowded. In hindsight, I should have investigated the behavior more deeply and continued advocating for the solution.