Decision Fatigue in Business: For Founders, the Cost Is Re-Deciding

The number everyone quotes is that an adult makes about 35,000 decisions a day. It gets used to explain why owners are exhausted by Thursday, and it is the wrong explanation. Almost all of those 35,000 are free. You pick a sandwich, you choose a font, you route a delivery, and the choice is over the moment it is made. It never comes back.

The decisions that wear owners down behave in the opposite way. There are very few of them — in most businesses, four or five at any one time — and they are never over. They get raised, thought about, half-answered, and left open. Then they come back on Sunday night, and in the truck, and at eleven at night, and each time you do the entire piece of thinking again from the beginning.

That is the cost. Not volume. Repetition.

What Decision Fatigue in Business Actually Looks Like for Founders

The clinical version of decision fatigue is real and well-documented: across a long sequence of choices, quality degrades, and late in the day people start defaulting to whatever takes the least effort. But in a small business the label gets applied so broadly that it stops being useful. An owner says they have decision fatigue and means they are tired, and the advice they get back is about batching email.

The version that actually costs money is much narrower, and you can identify it by a single test: can you name, out loud, what you decided? If you can, the decision is closed, and it is not costing you anything to carry. If you find yourself saying "well, I'm leaning toward" — it is open, and it will be back within the week.

Open decisions have a distinctive set of symptoms, and none of them look like tiredness. You go quiet on a question every time it comes up in conversation. You ask for one more piece of information you do not need and will not read. You notice the same topic appearing in your notes app in three different months. You have an opinion you have held privately for weeks and have never said to anyone who could act on it.

An Open Decision Keeps Charging the Old Price

Here is the part that makes this expensive rather than merely annoying. While a decision is open, the business does not pause and wait for you. It keeps running on the last decision you made — which means an unmade decision is a decision to keep doing the current thing, taken silently, every single day, without anyone noticing they took it.

Work an illustrative case through, with round numbers chosen to make the mechanism visible rather than to describe any real business. A service company quotes jobs at an average of $2,400. In June the owner works out that the rate is low and should go up by about 8%. They do not close the decision. They think about it in July, discuss it in August, and finally raise the rate in October.

In the four months in between the company quoted twenty-two jobs. Eight per cent of $2,400 is $192. Twenty-two jobs at $192 is $4,224, and that money was not lost to a bad decision. There was never a bad decision. It was lost to a correct decision that stayed open, and every week it stayed open it re-authorised the old price.

Run the same arithmetic on the time. If each of those three or four open questions genuinely surfaces three times a week, and each resurfacing takes ten minutes of real attention — a conservative figure, since a decision you are avoiding tends to take the good ten minutes rather than the tired ones — that is about ninety minutes a week spent deciding things that do not get decided. Across a quarter it is roughly nineteen hours. The output of those nineteen hours is nothing, because none of it ended in a sentence anyone could read back.

You cannot be fatigued by a decision you have closed. You can only be fatigued by one you keep reopening.

A Short Exchange at the Board Table

An owner brought a version of this to a session — three decisions open since the early summer, one of them a rate increase they were fairly sure about. Two advisors took it, and they did not agree about what the problem was.

Robert "Bob" Callahan, Business Fundamentals Advisor

The useful thing you did was count the months. Most owners never do, which is exactly why this goes on for a year instead of a quarter.

Let me ask you something simple. Of those three, which one changes what a customer pays you? Because that is not really open. You have been charging the old rate every week since June, and you chose that. Not deliberately, but you chose it — you just did it by not saying anything. There is no neutral position on a price. There is the price you are charging.

I watched this happen in my own business with a materials surcharge for the better part of a year. I told myself I was being careful. I was being slow, and slow cost more than being wrong would have, because being wrong I would have found out in a month and corrected it. Being slow I found out in eleven.

So here is the boring question. If you raise the rate this week and two customers leave, are they customers who come back every year, or are they the ones who shop you against three quotes every time? If you cannot answer that, that is the decision you are actually stuck on, and it is a smaller one than the one you thought you had.

Lisa Gomez, People & Culture Advisor

You are tracking your own patterns honestly, which is rarer than it sounds, and it is the thing that makes this fixable.

But I want to name what I see, because I do not think closing these three faster solves your problem. Three decisions stayed open for four months and there was never a point at which anybody other than you could have closed any of them. That is the finding. Next quarter there will be three more, and they will sit for four months too, because the queue is not the issue — the fact that it has exactly one server is the issue.

Your team knows this about you, by the way. It is worth sitting with. People stop bringing you things they could have handled, not because they are lazy, but because they have learned that everything routes through you and the queue is long. So they wait. And then you experience the waiting as proof that you have to decide everything, which is how this hardens.

The fix is a decision-rights conversation, not a productivity system. Name one category and one dollar figure — under this amount, in this area, somebody else decides and tells you afterwards. You will hate the first three outcomes. Two of them will be fine.

So: who on your team has ever made a decision you disagreed with and kept their job comfortably? If nobody comes to mind quickly, that is your answer about why the queue looks like it does.

Where They Landed Differently

Bob's position is that the three decisions are the problem and they should be closed this week, because each week they stay open has a price attached that the owner is already paying. Lisa's position is that closing them changes nothing that matters, because the mechanism that let three decisions sit for four months is untouched and will simply produce three more.

Both are right about different time horizons, and the owner has to pay for both: close the open ones now, because they are costing money now, and fix the routing, because otherwise this is an annual event. What Evelyn Hart's minutes do with a disagreement like that is not resolve it. They record it, with a date against each half, so that at the next session the first question is the one she always asks — this was discussed last week; what changed?

What Closing a Decision Actually Requires

A decision is closed when four things are written down. Owners reliably write the first three and skip the fourth, and the fourth is the one doing the work.

1. What was decided, in one sentence, in the active voice. "Rates go to $2,590 on standard installs."
2. Who owns it. One name. Not a department, not "we."
3. When it takes effect. A date, not "soon" and not "next quarter."
4. What would reopen it. The specific, observable condition under which this decision gets revisited — and nothing else does.

That fourth line is the anti-relitigation device, and it is worth understanding why it works. Without it, every doubt is equally valid as a reason to reopen the question, because there is no standard to test a doubt against. A bad week, an offhand comment from a customer, a competitor's website — any of it can restart the whole argument at eleven at night. With a stated reversal condition, you have something to measure the doubt against. "Reopen if we lose two repeat customers in a quarter specifically over price" means that one irritated customer on a Tuesday is not a reason to reopen anything. You can set the thought down without having to resolve it.

This is exactly the shape of a Boule Record: the decision in one line, the owner, the date, and whatever was left unresolved carried forward explicitly rather than silently. The reason to write it somewhere outside your own head is that your head is the one place where "we sort of decided" and "we decided" are indistinguishable.

The Sort That Comes First

None of the above means every decision deserves a written record. Most deserve the opposite — a rule that makes the decision once and then never again.

Sort by two things only: how hard it is to reverse, and how far it reaches. Pricing, hiring, leases, a new market, a partnership — hard to undo, reaches everything. These get your fresh attention, outside input, and the four lines above. Software subscriptions, supply orders, scheduling, response timing — cheap to reverse, reaches almost nothing. These should never be decided individually. They should be governed by a standing rule: any expense under $200 in this category is pre-approved; any request that adds more than two hours gets quoted separately.

The test for whether something has become a rule is whether you can say it out loud in one sentence. If you cannot, it is still a decision, and it will come back to you every time it arises — which is precisely how a category of genuinely trivial choices ends up occupying the same queue as the rate increase.

If the deeper problem is that you are the only server on that queue, the fix is a delegation problem rather than a discipline problem, and it is worth treating as one. If the problem is which of the open items to close first, the sequencing question is its own piece of work. And if you want the mechanical version of what a standing group of advisors does with items like these week to week, that is here.

One Thing to Do This Week

Take a sheet of paper and write down every decision currently open in your business. Most owners expect twenty and find between four and seven. Next to each one, write the date it first came up. The ones more than a month old are the entire problem, and one of them is almost certainly charging you the old price right now.

Close one of them before Friday. Four lines. It does not have to be the right decision; it has to be a closed one, with a written condition that would tell you it was wrong.