Test the decision before you commit to it

A decision feels most certain at the moment it is made. The meeting has ended, the argument is over, and everyone wants to move. That is also the cheapest moment there will ever be to find out that it is wrong.

Once a business acts, the cost of changing course climbs. People are hired or moved. Suppliers are signed. Customers are told. Each step makes the decision harder to reverse, and each step adds someone with a reason to defend it. A month on, nobody asks whether the decision was right. They ask how to make it work.

So the useful time to test a decision is before anyone commits to it. Testing is not delay. It is a short, deliberate look at what the decision rests on, made while changing your mind still costs nothing.

That is also the cheapest moment there will ever be to find out that it is wrong.

What a decision rests on

Every decision rests on a handful of beliefs. Some are about the market. Some are about the business itself, such as how much capacity it has, how long a change will take, or how a key person will react. Some are about timing.

Most of these beliefs are reasonable. A few are guesses that have been repeated so often they now sound like facts. The work of testing is to tell the two apart.

Start by finishing this sentence. "This decision is right if..." Write down every ending you can think of. Then ask, for each one, how you know. Some answers point to evidence you can go and look at. Others point to one person's opinion, a past experience, or a figure nobody has checked lately. Those are the ones to test.

How to test it

Testing does not need to be elaborate. For most decisions it is three kinds of work.

Read what the business already knows. The management accounts, the customer records, and the reports the board sees each month often hold the answer to a question nobody has put to them.

Go and look. Time with the people who do the work shows how the business runs, which is not always how it is described in the boardroom. A process that looks simple on paper can depend on one person's memory. A team that looks stretched can turn out to be waiting on another team.

Ask the people who will carry out the decision what would stop it working. They usually know. They are rarely asked before the decision is made.

What testing changes

Sometimes testing confirms the decision. That is a good result. The business can act with more confidence, and it knows which assumptions to watch as the plan unfolds.

Sometimes testing changes the decision a little. The timing moves. The order of steps changes. A risk nobody had named is now in the plan with someone responsible for it. These changes are cheap before the decision and expensive after it.

Occasionally testing shows that the decision should not be made at all. This is the hardest result to accept and the most valuable. The cost of finding out now is a few weeks of work and some disappointment. The cost of finding out later is the decision itself.

Why it is skipped

If testing is so useful, why is it skipped so often? Three reasons come up again and again.

Momentum. Once a leadership team has agreed a direction, raising a doubt feels disloyal or slow. Testing looks like a lack of nerve.

Familiarity. The people closest to a decision know the business well, and that knowledge makes it easy to assume that what was true last year is still true.

Time. A decision that has to be made this quarter does not feel as if it has room for checking.

Each reason is understandable. None of them makes the decision safer.

A habit, not an event

Testing works best as a routine. For every significant decision, ask what it rests on and how the business knows. Expect the answer to be written down. And treat a decision that changes after testing as a sign that the routine is working, not as a failure.

A business that tests its decisions before it commits to them will still make mistakes. It will make fewer, and it will find them sooner. That is the whole point.

It is also how we work. We test the facts of a decision before we advise on it, because that is when testing is worth the most.