The Decision-Latency Tax Most Growing Businesses Never Measure

Every business tracks whether decisions eventually get made and, roughly, whether they turn out well. Almost none track a separate, quieter number: how long each decision sits waiting for approval before it actually happens. That waiting time has a real cost attached to it, independent of whether the eventual decision was good or bad, and it's a cost most businesses have never once measured, let alone tried to actively reduce.

This gap is worth taking seriously because decision latency isn't a minor operational quirk — it's a tax applied to every piece of work that depends on a decision, charged in the currency of lost time, and most businesses are paying it continuously without ever having calculated the total bill.

Why decision speed and decision quality get bundled together

A decision that took a long time and turned out well looks, in hindsight, like it was worth the wait — which makes the actual cost of the waiting itself nearly invisible in any retrospective.

When a delayed decision eventually turns out fine, the delay tends to disappear from the story entirely — nobody remembers to ask whether the outcome would have been even better with less time lost. Decision quality dominates the conversation after the fact, and decision speed quietly exits the discussion, even though speed had its own real, measurable cost the whole time, separate from and in addition to whatever quality the eventual decision turned out to have.

What the actual cost of waiting looks like

Every day a decision sits waiting is a day whatever depended on that decision also sits waiting — a project paused, a hire delayed, a deal that can't move until an answer arrives.

The cost of decision latency isn't abstract — it's every downstream activity that depended on that decision sitting idle in the meantime. A pricing exception that waits three days for approval holds up a deal for three days. A hiring decision that waits two weeks for sign-off delays everything that new hire was going to take off someone else's plate for two weeks. Multiply that across every decision in a growing business, and the idle time adds up to something substantial, even though no single instance of it ever looks dramatic enough to draw real attention on its own.

A concrete picture of how this latency compounds across a single week

A business handling forty decisions a week, each averaging two days of waiting before approval, is carrying roughly eighty person-days of idle, downstream time every single week — a number nobody is currently tracking.

Picture a growing business processing roughly forty decisions a week that require some form of approval — pricing exceptions, scheduling conflicts, small resource allocations. If each of those decisions averages two days of waiting before it actually gets approved, the business is collectively carrying roughly eighty person-days worth of idle, downstream waiting time every single week, spread invisibly across dozens of different projects and people, none of which shows up as a single line item anywhere in the business's actual operating metrics.

Why this compounds specifically as a business grows

More people means more decisions routing through the same limited approval capacity — so the average wait time per decision tends to get worse with growth, not better, unless something changes about how decisions route.

As a business adds people, it also adds more decisions that need approval, while the number of people actually authorized to approve them often doesn't grow at the same rate. That mismatch means average decision latency tends to worsen with growth by default, quietly taxing a larger and larger share of the business's total activity the bigger it gets, unless someone deliberately intervenes to change how approval authority is distributed as the business scales.

Why this measurement is harder to run than it sounds

Capturing an accurate timestamp for when a decision was first requested, separate from when it was actually approved, requires a level of process discipline most businesses haven't built — which is part of why this measurement almost never happens organically.

Actually measuring decision latency requires capturing two separate timestamps for every decision — when it was first requested, and when it was actually resolved — which is a level of process discipline most businesses haven't built into their normal workflow. Without that discipline, the measurement simply can't happen, which is part of why this particular cost remains invisible even to businesses that are otherwise diligent about tracking their operational metrics closely.

What measuring this actually enables

Once decision latency has a real number attached to it, a business can target the specific decisions worth pushing down to someone closer to the work — the ones costing the most in waiting time, not just the most frequent ones.

Once this latency is actually tracked, it becomes possible to identify which specific categories of decisions are costing the most in cumulative waiting time, and deliberately push authority for those down to someone closer to the work. That's a far more targeted fix than a general instruction to "delegate more," because it's based on where the waiting is actually most expensive, not just where delegation feels intuitively overdue, and it gives a business a concrete, prioritized list to work through rather than a vague sense that things could probably move faster somewhere.