Every engineering organization knows the concept of technical debt: shortcuts taken in code today that someone will pay for later, with interest. Mature teams name it, track it, and budget time to pay it down. It's one of the most useful ideas software ever produced.
Companies carry an equivalent debt in their decisions. Nobody tracks it. Most haven't named it.
Decision debt is the accumulated cost of decisions that were never properly made, never truly owned, or never written down. Each one seems cheap to defer. Together, they quietly become one of the largest line items in the company — paid in the most expensive currency available: senior attention.
The three forms it takes
Deferred decisions — never made. The pricing policy that's been "almost ready" for three quarters. The market entry that's perpetually "under evaluation." The org question everyone knows must be answered and nobody wants to answer. While the decision waits, it doesn't sit still: everyone routes around the gap, each in their own direction. Sales improvises one answer, delivery improvises another, and by the time the decision is finally made, it has to overturn six months of improvised precedent — which makes it even harder to make, which extends the deferral. Deferred decisions don't wait. They breed.
Orphaned decisions — made, but not owned. A meeting happens. Heads nod. Something is "decided." But no single role owns the decision, so nothing enforces it. Within weeks it gets reopened — usually by whoever feels most strongly that day, or whoever wasn't in the room. The company believes it decides things; what it actually produces are temporary opinions with attendance lists.
Unrecorded decisions — made, but not written. The decision lives in the memory of the people who were there. New hires can't find it, so they relitigate it, in good faith, one by one. Institutional memory becomes three veterans who are tired of explaining — and a single resignation away from amnesia.
The interest payments
Decision debt is invisible on every dashboard, but you can hear it if you know the sound:
The same topic resurfacing in every quarterly meeting, slightly rephrased. "Alignment meetings" about things that were decided months ago. A calendar full of discussions that produce discussion. New hires asking why something works this way and receiving, instead of a reason, a history lecture. Decisions made by stamina — whoever can attend the most meetings and repeat their position the longest wins.
Each individual unmade decision costs little per day. But the costs compound, because every gap forces workaround decisions, and the workarounds conflict, and resolving the conflicts requires more meetings — which is how a company arrives at the strange place where half its calendar exists to service old decision debt, and everyone is too busy aligning to notice.
Why it accumulates
Three forces, all of them rational at the individual level.
First, the asymmetry of risk. Making a decision attaches your name to an outcome. Deferring it attaches your name to nothing. In most companies, the punishment for a wrong decision is vivid and the punishment for a slow one is invisible — so people respond to the incentives exactly as designed.
Second, consensus culture. In organizations where every stakeholder can softly veto, the price of deciding is a negotiation marathon, and the cheap alternative is to keep "gathering input." Consensus cultures don't lack decisions because people are weak; they lack them because the system makes deciding expensive and deferring free.
Third, growth. Decision rights are usually defined — implicitly — for the company's size at some earlier moment. The rules written at fifty people are still silently in force at three hundred. Authority never moved, so everything important still flows to the same few calendars, and the queue outside those calendars is decision debt, accumulating daily.
Paying it down
The good news: decision debt responds quickly to embarrassingly simple treatment.
- Take inventory. Ask one question across the leadership team: which decisions have we made more than twice in the last twelve months? The list writes itself, and it's the debt register.
- Assign owners — roles, not names. For each recurring decision: one role decides, others are consulted. The sentence that does the work is short: "[Role] decides [X]. Everyone else gets consulted, not asked for permission." It feels almost too small to matter. It changes everything downstream.
- Write decisions down. One page or less: what was decided, by whom, when, and when it will be revisited. The format must be light enough that it actually gets used — a heavyweight decision-log process is just a new way to defer.
- Recommit or kill, on a cadence. Once a quarter, the oldest open decisions either get made or get explicitly parked with a date. "We are not deciding this until March" is a decision. "We should discuss this again sometime" is debt.
Measuring it
If you want numbers — and this site exists to argue that you should — decision debt is one of the most measurable things in the human layer. Count reopened decisions per quarter. Track lead time on the ten most important recurring calls: raised to resolved. Estimate the share of meeting hours spent on previously decided topics. None of this requires software. It requires a tally and the willingness to look.
And here's the practical magic: even rough counting changes behavior. The moment "we reopened eleven decisions this quarter" exists as a sentence, reopening the twelfth costs something socially. Measurement isn't only diagnosis. At this layer, it's already treatment.
Companies audit their financial debt every year, to the cent. Decision debt appears on no balance sheet, yet plenty of organizations pay more interest on it than on anything the bank holds.
What's the oldest unmade decision in your company? That's where I'd start counting.