Why We Asked So Many Questions Before Rebuilding Atlas CRM

If you saw our last blog, you know we've been taking Atlas CRM apart to build something better: new foundations, new navigation, dozens of screens reworked. That post showed what is changing. This one explains how we got there.

Quick refresher: Forge is Atlassian's newer platform for building apps, the one that plugs into the future of Atlassian, including Rovo and the Teamwork Graph. Sounds technical and boring, until you realize it's what lets Atlas CRM keep up with where your tools are heading.

The Forge version of Atlas CRM is in beta now. If your admin has turned it on, you can already see the result of everything below. If not, this is what we've been working on, and why. And we wanted to get it right.

So we asked. A lot.

Research is discovery

A lot of teams use research to confirm what they've already decided. We use it to figure out what to decide in the first place.

The Forge rebuild touches almost every part of the product. Navigation, layout, the way customer data flows between Jira and Confluence, all of it back on the table. Without input from the people actually using Atlas CRM, we'd just be redecorating the same room with our own taste.

So before we touched the design, we set up three layers of research.

First, the panel. Real Atlas CRM users who signed up to help shape the product. These are the people who use Atlas CRM every day, came to a demo, or reached out to support.

Second, one-on-one product chats. Regular conversations where we walk through actual workflows with customers. Not "what do you want?" but "show us what you're trying to do." Those sessions surface things a survey never will.

Third, prototype testing. Before we wrote production code, we put the new designs in front of users and watched what happened. Did the navigation make sense? Could people find what they needed? We tested it instead of assuming it.

We tested the new navigation with real users before writing a single line of production code. Every participant found it consistent with Jira and Confluence. First impressions averaged 4.3 out of 5. Words like "familiar," "clear," and "easy to navigate" came up again and again. That told us the direction was right.

Not everything landed perfectly. Several users asked for a home screen that adapts to their role, because what a sales rep needs to see first thing Monday morning isn't what a support team needs. Others want deeper reporting. Some felt the design leaned too close to standard Jira and lost a bit of Atlas CRM's identity. That's exactly why we test early. We're working on it.

Each layer catches different things. The panel tells us what people need. The chats show us how they work. The prototype tests tell us whether our solutions actually land.

Every week, our product trio — product, design, engineering — sits down with what we've heard and figures out what to do about it. Marketing and business development get pulled in too, because feedback that only reaches the product team is feedback half-wasted.

The moment Atlas CRM clicks

Every Atlas CRM user has a moment when the product clicks. Through continuous interviews, we've discovered these moments follow clear patterns.

One user imported their customer data and immediately showed their CEO how everything connected. Another realized they could link customer fields directly into Jira workflows. A third saw their scattered customer information unified in one place for the first time.

These moments reveal what drives adoption. Not the features we think are impressive, but the specific workflows that solve real problems. And those are exactly the moments we're protecting in the Forge version.

The most valuable insights come from watching how customers adapt Atlas CRM to their workflows. Instead of just telling us what they want, they show us what they're trying to accomplish.

One customer built their own interface for bulk data management because they needed capabilities we didn't offer natively. Instead of just noting the gap, we explored why their approach worked and how those patterns could improve Atlas CRM for everyone.

This continuous dialogue generates ideas we never would have considered internally. That conversation is going into the Forge version.

Testing on Forge, in real time

Forge changed the speed of all this.

Research prevents us from building features that sound good but don't create real value. When we understand what drives genuine adoption, we can focus resources on workflows that matter most.

The data shows clear patterns about where users experience value and where they hit friction. Instead of building every requested feature, we're optimizing the path to those critical moments and removing obstacles that prevent users from reaching them.

If we were doing this in a vacuum, we'd be shipping someone else's CRM. Not yours.

The customer panel gives us early signals about market direction and validates new concepts before significant development investment. Customers who help shape the product become advocates who can articulate value to their organizations more effectively.

What's coming soon

The new Atlas CRM lands soon. New navigation, sharper workflows, dozens of screens reworked, all shaped by what users told us. It looks a bit different. It feels familiar.

Under the hood, it's now built on Forge, which means we're ready for what comes next in Atlassian. And here's the part that matters most if you've read this far: your data stays where it is. Your configurations and settings carry over. Automations keep running. The API doesn't change. There's no downtime planned.

See you soon.