The coordination tax

Asana surveyed over 10,000 knowledge workers and found that 60% of the average knowledge worker's day is consumed by coordination: communicating about work, searching for information, switching between apps, managing shifting priorities, and chasing status updates. Only 27% goes to the skilled work they were hired for. The remaining 13% goes to strategic planning.

The breakdown in hours is even more striking. Over the course of a year, the average knowledge worker spends 103 hours in unnecessary meetings, 209 hours on duplicative work, and 352 hours communicating about work rather than doing it. That's 664 hours, roughly 16 full working weeks, spent on the overhead of work rather than the substance of it.

These numbers are not anomalies from one survey. Microsoft's Work Trend Index found that users spend 60% of their time in Microsoft 365 on emails, chats, and meetings, and only 40% in creation tools like Word, Excel, and PowerPoint. Harvard Business Review researchers found that time spent on collaborative activities has increased by 50% or more over the past two decades, with many employees now spending 80-85% of their time on coordination.

The coordination tax is real, it's growing, and it's structural. Understanding what it's made of is the first step toward reducing it.

The four components

The coordination tax breaks down into four categories, each with its own root cause and its own fix.

Status coordination

Time spent figuring out what's happening: standups, status meetings, "just checking in" messages, project updates, weekly syncs. Each one exists because the current state of work isn't visible to the people who need to see it.

The root cause is that status lives in people's heads rather than in a system. The engineer knows their PR is blocked. The project manager knows the timeline has shifted. The designer knows the mockup is ready for review. But none of this is visible to anyone else until someone asks or a meeting is held to share it.

The fix is making status visible without requiring synchronous communication. Self-writing documentation that tracks what's shipped, what's in progress, and what's blocked from GitHub activity and project conversations provides the same information as a standup without requiring everyone to be in the same room at the same time.

Context transfer

Time spent explaining background, sharing knowledge, and bringing people up to speed: onboarding walkthroughs, "let me give you some context" meetings, the first twenty minutes of every discussion that's spent establishing shared understanding before the actual topic can be addressed.

The root cause is undocumented institutional knowledge. The context exists but it's locked in people's heads, which means sharing it requires the knowledge holder to be present, available, and willing to repeat themselves.

The fix is documentation that captures context as it's created. When decisions are documented with their reasoning, when system architecture is described in searchable docs, when meeting discussions are captured and indexed, the context is available on demand without requiring anyone to explain it live.

Decision reconstruction

Time spent figuring out what was decided, who decided it, and why: relitigated discussions, "didn't we already decide this?" conversations, meetings called to re-examine questions that were resolved months ago but whose resolution was never documented.

The root cause is that decisions aren't documented. The discussion happened in a Slack thread that's scrolled away, a meeting that wasn't recorded, or a conversation between two people that nobody else was part of. The decision was made but the record of the decision and its reasoning disappeared.

The fix is decision logs that capture what was decided, who was involved, what alternatives were considered, and why the chosen approach was selected. These can be generated automatically from Slack discussions and meeting recordings where decisions are actually made.

Search overhead

Time spent looking for information that exists but can't be found: searching across multiple tools, asking "does anyone know where..." in Slack, browsing folder structures hoping to recognise the right file.

The root cause is information fragmentation. The information exists but it's spread across tools that don't share context. Employees waste 1.8 hours per day searching for things they already have.

The fix is a unified search layer that spans all tools and finds information by meaning rather than by keyword. When one search bar covers Slack, Google Drive, GitHub, email, and every other tool where knowledge lives, the search overhead drops dramatically.

Why culture can't fix this

The standard response to coordination overhead is cultural: "let's have fewer meetings," "let's communicate more efficiently," "let's be more disciplined about documentation." These appeals rarely produce lasting change because they're asking people to work against the grain of their tools and organisational structure.

When the only way to share status is a meeting, meetings will happen. When the only way to transfer context is a conversation, conversations will consume the day. When the only way to find information is to ask someone, interruptions will proliferate. The culture isn't the problem. The infrastructure is.

Reducing the coordination tax requires structural changes: documentation that writes itself from existing activity, search that works by meaning across all sources, decision logs that capture reasoning automatically, and async-first processes that make synchronous coordination the exception rather than the default.

The 60% figure is not a law of nature. It's a measure of how much coordination overhead the current infrastructure imposes. Change the infrastructure and the number changes with it.

Frequently asked questions

Is 60% really accurate? Multiple independent studies converge on similar figures. Asana's 60% is consistent with Microsoft's 60/40 split between coordination and creation tools. The specific percentage varies by role, industry, and company size, but the pattern is consistent: the majority of knowledge worker time goes to coordination rather than production.

Which component should we tackle first? Start with whichever one causes the most pain. For most teams, it's either search overhead (people can't find things) or context transfer (too many meetings exist to share knowledge). Both are addressed by making knowledge documented, searchable, and current.

Won't reducing coordination make teams less aligned? The goal is better coordination, not less coordination. Async documentation provides more complete and more persistent alignment than synchronous meetings, because the documentation reaches everyone (not just attendees) and persists (rather than fading from memory within days).

Related reading: Work about work, The knowledge scaling problem, Why meetings multiply, The cost of scattered knowledge. Related pages: Self-writing docs, One search, Decision log.