The Handoff Tax

An engineer ships a fix for a billing edge case. Two weeks later, a designer proposes a flow that reintroduces the exact same edge case. Nobody notices, because the original context, why the fix was needed, who it affected, what the failure looked like, never traveled with the work. The decision was made, solved, and then quietly forgotten. That cost has a name. It's the handoff tax.

Every product gets built through a chain of translations. A user hits a wall and tells support. Support summarizes it for a product manager. The PM reframes it for a roadmap. Design turns the roadmap into screens. Engineering turns the screens into code. Leadership turns the launch into a slide. Each step is reasonable. Each step is performed by competent, well-meaning people. And each step is a compression, because the only way to move work between groups is to package it into something that travels, and the things that travel are summaries, tickets, and artifacts.

Compression isn't evil, it's necessary. But it always loses something, and the parts it loses are the parts that mattered most, because the parts that matter are usually the hardest to summarize. The situation. The stakes. The constraint nobody wrote down. The reason the obvious solution is wrong. By the time the work reaches the person who will actually build it, the user who started the whole thing is gone, replaced by a mascot stitched together from assumptions.

You can watch this happen inside a single ticket. Someone files "Budgeting UX improvements," description: "Users reported issues with the budgeting workflow. Design to simplify." That sentence is perfectly legible to the system and almost useless to anyone trying to understand what the user was actually fighting. Six weeks later, someone reads it and starts making decisions based on the word "simplify." Simplify by removing fields? By hiding options? By cutting the advanced features the largest accounts depend on? All of those are simplifications. None of them are necessarily right, and there's no longer any way to know, because the moment that would have told you got compressed away three handoffs ago.

The instinct, when you see this, is to call for more empathy. Tell the team to care more about users. That doesn't work, because the problem was never a shortage of caring. The people doing the translation are full of smart, well-meaning humans. The problem is structural. Organizations reward legibility, speed, and smoothness. Handoffs feel like progress. Artifacts that can be moved between groups feel safer than ambiguity. So the system keeps translating away the one thing that actually matters: the user's lived moment.

The handoff tax is not a people problem. It's a boundary problem. Every boundary, person to person, team to team, this quarter's plan to next quarter's, compresses the context, and every compression drops the parts that don't travel. You can remove almost every human from the chain and the tax still walks back in, because the boundaries are still there.

So the fix isn't to try harder. It's to change what's hard. Right now, losing context is easy and keeping it is heroic. You want to flip that, so that context is harder to lose than to keep. That means receipts instead of summaries: a real clip, a real quote, a real moment attached to the work, specific enough that it can't be quietly reinterpreted into something safer. It means decisions that are retrievable, so the reasoning behind a choice survives the person who made it. It means letting reality interrupt the plan early, while it's still cheap to change course, instead of discovering at launch that you solved a problem nobody had.

None of this is about slowing down. It's the opposite. The handoff tax is what makes teams slow, because they spend their energy rebuilding context that was already paid for once, relitigating decisions that were already made, and shipping fixes that get undone by the next person who didn't have the receipts. A team that keeps its context moves faster, because it isn't constantly starting over.

The first time you really see the tax, you can't unsee it. You start noticing the moment in every meeting when the user slips out of the room, when a vivid, specific problem becomes a generic theme that everyone can nod at and no one can act on precisely. The work isn't to eliminate handoffs. It's to build so that what matters survives them.

This is the argument of my book, Intent: How to Build Products That Last in the AI Era.