How to Make Decisions Faster as You Scale
Something happens to founders’ personal decision-making as companies grow: calls that used to take an afternoon now sit open for weeks. The usual explanation is that the decisions are harder, and that is partly true. The larger part is that the consequences got bigger, more people are watching, and being wrong now costs something visible — so the rational response is to gather more information, which feels like diligence and is frequently just delay. This page is about the personal practice, as distinct from the organisational sources of slowness.
What’s inside
Why your decisions slowed
Sort by reversibility first
The good-enough threshold
Four reasons a decision is stuck
Deciding with dissent in the room
The decision journal
Where Blomma fits
Why your decisions slowed
Three things changed, and none of them is that you got worse at deciding.
The consequences got bigger.A wrong call at ten people cost a week. At a hundred and fifty it can cost a quarter and a few good people. That asymmetry makes caution rational, and caution shades into delay without any moment where it becomes obviously excessive.
You have less direct information.Early on you decided from first-hand knowledge. Now the information arrives through layers, already interpreted, and you know it is incomplete — so the instinct is to keep asking, which is often correct and often an infinite loop.
You are being watched.Your decisions are read as signals, and a decision that turns out badly is remembered. That adds a cost to being wrong that has nothing to do with the decision’s actual stakes, and it pushes toward the option that is easiest to defend rather than the one most likely to be right.
Understanding these three matters because they point at different fixes. The first calls for sorting by reversibility. The second calls for accepting a threshold of information rather than pursuing completeness. The third calls for noticing when you are optimising for defensibility.
Sort by reversibility first
The single highest-leverage habit, and it resolves most of the problem on its own.
For every open decision, ask how hard it is to undo. Reversible decisions — a pricing experiment on one segment, a process change, most hiring at non-senior levels, a marketing bet — should be made fast, on incomplete information, by whoever is closest. The cost of being wrong is bounded; the cost of delay is not.
Irreversible or expensive-to-undo decisions — a market entry, a platform commitment, a senior hire, a fundraise structure, a layoff — warrant real deliberation. Slowness here is not latency, it is appropriate.
The failure most founders make is applying uniform care to both, which means they are simultaneously too slow on the reversible majority and insufficiently careful on the irreversible minority. Sorting your open decisions this way usually reveals that most of what is stuck is reversible, which means it should simply be decided today.
One refinement: check whether a decision that looks irreversible has a cheap reversible version. Frequently the right move is not to make the big call but to buy the option — a pilot, a contractor instead of a hire, a limited launch.
The good-enough threshold
Deciding requires accepting that you will never have complete information, which sounds obvious and is the thing that actually blocks people.
A usable threshold: decide when additional information is unlikely to change the answer. Not when you are confident — when you can no longer name a specific piece of information that would flip your view. If you cannot name it, you are not gathering, you are hesitating.
That test is useful because it is concrete. Ask yourself what specifically you are waiting for. “More clarity” is not an answer. “The March numbers” is, and it tells you whether waiting is justified and exactly how long it should take.
Two further prompts.What would I do if I had to decide today?Usually you know, and the answer is the same one you will reach in three weeks.What is the cost of the delay?Founders account carefully for the cost of being wrong and almost never for the cost of the decision sitting open, which includes everyone downstream waiting.
Four reasons a decision is stuck
Most stuck decisions are not information problems. Four causes, with different unblocking moves.
Genuinely missing information, obtainable soon.The only case where waiting is right. Name the information and the date, and put it in the calendar rather than leaving the decision open-ended.
You do not want either option.The decision is between two unattractive outcomes, and delay feels like a third option. It is not — delay is choosing the worse of the two with extra cost. The unblocking move is naming that explicitly: “both of these are bad, and not choosing is the worst one.”
You are avoiding the conversation the decision requires.Extremely common. The decision is clear and executing it means telling someone something difficult. This presents as indecision and is actually avoidance. The unblocking move is to admit which conversation you are avoiding and prepare it.
It is not actually your decision.You are holding something that belongs to someone else, so you cannot resolve it well — you lack the context — and neither can they. The unblocking move is to hand it over.
Run these four on anything that has been open more than two weeks. In most cases the answer is the second or third, and neither is fixed by more analysis.
Deciding with dissent in the room
A specific skill at scale: making a call when capable people disagree.
The mistake is waiting for consensus. On genuinely hard decisions consensus either does not arrive or arrives because people stopped arguing, and a decision made because everyone got tired is not better for the delay.
A workable sequence. Make sure the disagreement is understood — have each side state the other’s position well enough that the other accepts it. That step frequently resolves the disagreement, and where it does not, it identifies the actual crux. Then decide, and say what you weighed. Then ask explicitly for commitment: not agreement, commitment — executing it as though it were theirs, including in front of their own teams.
And record the dissent. Naming that someone disagreed and why does two things: it means they do not have to keep re-raising it to feel heard, which is the main cause of decisions unravelling, and it gives you something honest to review against later.
What undermines this: deciding without acknowledging that you heard the objection. People will accept losing a decision considerably more easily than they will accept having their reasoning ignored.
The decision journal
A cheap practice with a compounding return, and almost nobody does it.
For significant decisions, write three lines when you make it: what you decided, what you expected to happen, and what would tell you it was wrong. Ten minutes.
Three things follow. You get calibrated — most people are surprised, on review, at which kinds of decision they are reliably good and bad at, and there is no way to learn that from memory because memory reconstructs the reasoning after the fact. You stop re-deciding, because the reasoning is recorded and a new data point can be checked against it rather than reopening everything. And you get an honest record when someone — a board member, a cofounder, yourself — later characterises what you were thinking.
Review quarterly. The pattern that emerges is usually more useful than any framework: most founders find they are too slow on a specific class of reversible decision, and too fast on a specific class of people decision.
Where Blomma fits
The hard part of this is diagnosis, and it is hard because two of the four stuck-reasons are uncomfortable to admit. Recognising that you are not gathering information but avoiding a conversation, or that you are holding a decision that is not yours, requires a candour that is difficult to reach alone and impossible to reach with the people involved.
Blomma is an always-on AI career coach with no stake in what you decide. Use it to run your open decisions through the reversibility sort, which usually shows that most of what is stuck should simply be decided today. Use it to work out which of the four reasons applies to the one that has been open longest — the answer is rarely the first. Use it to prepare the conversation, if that turns out to be what you are actually avoiding. And use it to set up the decision journal and review it, which is the practice with the largest long-run effect on decision quality and the one nobody sustains alone.
The second application: the reason many decisions reach you at all is that people below you are stuck for the same four reasons and have nowhere to work it out. Coaching across a leadership team reduces the number of decisions that need you..
Your decisions did not slow because you lost your judgment. They slowed because being wrong got expensive and being late got invisible — and the fix is to sort by reversibility, name what you are waiting for, and be honest about which decisions are actually conversations.