Ever wonder why your engineering team is always busy but you're never shipping quite fast enough?
It's not because your engineers are slow. In fact, the average engineer today ships more code in a week than they did in a quarter five years ago. The bottleneck is everything that happens between the work: the tool switches and the 10 minutes spent figuring out which Jira ticket reflects the current spec, which version of the doc people are actually building from, and which Slack thread has the latest status update.
None of that shows up on a sprint board. But it shows up in your delivery dates.
This is the coordination tax that most engineering and product teams are paying. Technical work is inherently multisystem—our research found that 65% of technical workers use six or more tools in a typical week, few of which were designed to share context with each other.
Nearly two-thirds of technical workers say their tools feel fragmented rather than part of one unified system. And 61% say deciding how to do their work takes as much effort as actually doing it. That means the majority of your team's decision bandwidth is spent on navigating work, not executing it. This single data point encapsulates the coordination tax.
What the coordination tax costs
McKinsey's research found that developers spend only 32% of their time writing code. The remaining 68% goes to meetings, interruptions, and administrative overhead.
The cost compounds in ways that are easy to miss:
Every tool switch requires reconstructing context.
Every handoff between teams is a potential point of information loss.
Every manual status update is time not spent building.
Every clarifying question is a sign that context didn't travel with the work.
And here's where it gets counterintuitive: Adding more tools tends to make this worse, not better. More tools mean more surfaces to maintain, more places for context to fragment, and more decisions about where work should live.
Why AI isn’t fixing it … yet
AI was supposed to be the productivity unlock engineering teams had been waiting for. But for most, delivery still feels the same.
Our research points to why. We surveyed 478 technical professionals across engineering, product, design, and IT and found that the impact of AI on workload depends almost entirely on the environment it sits in.
In low-friction environments, where tools work together and context flows between systems, 51% of technical workers say AI reduces their workload. In high-friction environments, where tools don’t connect and keeping work moving requires constant manual effort, that flips: 50% say AI instead adds to their workload .
AI doesn’t fix a fragmented environment. It inherits it.
When your team has to manually bridge disconnected systems just to complete a task, adding an AI assistant to each of those systems adds another layer of context switching. The AI is only as useful as the context it has access to, and in a fragmented stack, that context is always incomplete.
What technical workers want
The research makes the contrast clear. In low-friction environments, the coordination tax drops sharply. AI starts doing what it was supposed to do. Teams spend more time building and less time coordinating.
The difference isn’t the quality of the people or the sophistication of individual tools. It’s whether those tools were built to work together. This is exactly what technical teams are asking for:
35% of technical workers say what they need most is tools that work better together.
32% want AI that understands their context and workflow.
31% want AI that works across the tools they already use.
That's not a wish list for better point solutions. It's a description of a connected, low-friction environment.
The path forward starts with the right diagnosis
The coordination tax is real, measurable, and addressable … but only if you diagnose the right problem first. Adding headcount doesn’t reduce it. Adding process doesn’t fix it. Adding more AI tools into a fragmented environment only makes it worse.
What moves the needle is understanding how much friction your teams are carrying and whether the conditions exist for AI to deliver what you’re investing in.
That’s exactly what the AI Value Quadrant is built to answer. It maps your organization and shows you what to prioritize to move toward a low-friction environment where AI value compounds instead of stalls.
