Gartner just published the first-ever Magic Quadrant™ for Developer Productivity Insight Platforms. Gartner estimates the DPIP market size at approximately $400 million with an average growth rate of over 40%, based on an assessment of global organizational spending on data-driven engineering analytics platforms. Swarmia was named as a Leader. I’m proud of that and feel it’s meaningful recognition for the team and for our customers.
From our point of view, the publishing of the report also marks an interesting shift in the market. I believe this is the first time Gartner has evaluated this space as a Magic Quadrant — previously it was a Market Guide, a format reserved for earlier-stage categories. The move signals that engineering intelligence has become a strategic priority, not just a nice-to-have. And I think the reasons behind that shift are worth talking about.
When I founded Swarmia in late 2019, the category barely had a name. Yet, Gartner predicts that “by 2028, 60% of developer productivity platforms will act as foundational context engines, equipping agentic workflows with real-time environmental awareness, state management, robust knowledge retrieval, policy guardrails, and strict goal alignment.”
These tools are becoming part of the infrastructure that makes AI-assisted development actually work. The reason the market is growing so fast is that AI has changed how software gets built, and nobody has figured out how to manage that change yet.
Steve Yegge’s piece on Anthropic’s ways of working has been on my mind since it came out in February. He describes a company where there's no roadmap beyond 90 days, where teams build around living prototypes, and where the work moves so fast that information from two hours ago is already stale.
Maybe that sounds extreme. But most engineering leaders I talk to are feeling some version of this. The old patterns are breaking down: six-week planning cycles, predictable sprint velocities, neatly scoped epics that move left to right across a Jira board. AI tools are making it possible for individual engineers to move much faster — but at the organizational level, the gains are still surprisingly modest for most.
That’s because writing code was only ever one step in a much longer chain. Think about a customer bug report. The code fix might be step seven of twelve. Before that, someone triaged the issue, someone reproduced it, someone figured out which team should take care of it. After the fix, someone reviews it, tests it, deploys it, and confirms it’s resolved. AI can make step seven happen in minutes instead of hours. But if steps one through six and eight through twelve stay the same, the customer still waits just as long.
This is what I mean by “optimizing the whole.” Most companies are still measuring how fast people type, when they should be measuring how fast value moves through the system.
Looking at how our customers work in 2026, I see three delivery modes emerging:
The roadmap. This hasn’t gone away. Quarterly planning, cross-team initiatives, milestone tracking — it’s still how most larger organizations coordinate. AI doesn’t eliminate the need for strategic alignment. If anything, it makes it more important, because teams can now move faster in the wrong direction too.
The campfire. This is what Yegge describes at Anthropic, and what both our teams and some of our most advanced customers are already doing. Small groups, prototyping together, immediate feedback loops. Instead of a heavy upfront spec, the team just starts building, and the end result is a living thing that evolves rapidly.
The stream. A constant flow of smaller improvements, fixes, and experiments. Less planned, more responsive to customer feedback and new, practical ideas. AI tools make this mode especially productive, because it’s great at making those small changes. But it also means a lot more throughput that needs to be visible and understood.
Most organizations won’t pick one of these. They’ll run all three at the same time, probably in the same team. And that’s going to be chaotic. Which is fine, but it means you’ll need even more visibility than you did before.
We built Swarmia around DORA, SPACE, and developer experience surveys. Those aren’t going anywhere. But I’d be lying if I said our current set of metrics fully captures what matters in a world where AI agents are writing pull requests and a single engineer can generate the output of three.
Here’s what I think engineering leaders will be dealing with over the next 12 months:
Value throughput should be going up — a lot. And I don’t just mean lines of code or PR count, but rather, something closer to how many stories and epics are moving through the system and reaching customers. If your teams have access to good AI tools and your throughput hasn’t changed, something is wrong. Either the tools aren't well-adopted, or the bottlenecks are elsewhere in the system.
The maintenance load will be growing fast. When generating new code is cheap, people generate a lot of it. But every line of code has a maintenance cost. CI costs are already spiking for some of our customers. More code means more builds, more merge conflicts, more things breaking. The internal tooling and infrastructure that nobody explicitly planned for is the part of the iceberg you can’t see. But even if you can’t see it, you still have to maintain it.
The spread between engineers is going to widen. Some engineers will get enormous value from AI tools. Others won’t — at least not yet. That creates new challenges for managers. How do you help someone who hasn’t found the right workflow? How do you make sure your most productive people aren’t creating bottlenecks for everyone else?
Usage stats from your AI vendor aren’t enough. You need to know whether the work is actually better, faster, cheaper — or just more. And you need that answer at the team and initiative level, not just as a company-wide average.
I have more to say about all of this, and we’ll be sharing more in the coming weeks and months — updated frameworks, new product capabilities, and a lot more stories from the field on how organizations of different sizes are living through these changes.
For now, if you want to understand where this market is and where it’s going, the Gartner report (accessible to Gartner subscribers only) is worth a read. It covers 15 vendors and gives a thorough view of what this space looks like today.
And if you want to talk about any of this — how your organization is adapting, what metrics still make sense, what you’re struggling with — send me a message. I mean that. This stuff is hard and nobody has all the answers yet. We’re all figuring it out together.
––
Gartner, Magic Quadrant for Developer Productivity Insight Platforms, Frank O’Connor, Peter Hyde, Akis Sklavounakis, Akriti Kapoor, 5 May 2026.
GARTNER and Magic Quadrant are a trademark of Gartner, Inc. and/or its affiliates.
Gartner does not endorse any vendor, product or service depicted in its research publications, and does not advise technology users to select only those vendors with the highest ratings or other designation. Gartner research publications consist of the opinions of Gartner's research organization and should not be construed as statements of fact. Gartner disclaims all warranties, expressed or implied, with respect to this research, including any warranties of merchantability or fitness for a particular purpose.
Subscribe to our newsletter
Get the latest product updates and #goodreads delivered to your inbox once a month.
