Ask what you would do differently rather than what went wrong, list three things, and write down what you would need to change for each. Skip the successes, which teach less.

The question that works

What went wrong invites defensiveness, and produces either a list of external causes or nothing at all.

What would you do differently is a different question. It assumes hindsight rather than fault, and people answer it honestly, including about themselves.

Ask it about the year and the answers come quickly, usually within a minute, because everybody already knows.

The purpose of the exercise is not discovering the answers. It is writing them down where they can be acted on rather than carried around.

Three, not everything

A constraint that makes the review useful rather than dispiriting.

A long list of regrets produces no change, because nothing on it feels achievable and the whole thing reads as failure.

Three items, chosen as the ones that cost the most, produce a January that is different from last January.

Choose by cost rather than by irritation, since the thing that annoyed you most is frequently not the thing that cost most.

Write the cost against each, roughly, in money or in hours, because that ordering is usually not the one instinct produces.

Where the answers usually are

The second is the most common and the most expensive. Deferred decisions cost the whole period they were deferred for, and that cost is invisible because nothing happened.

Separate the decision from the outcome

The distinction that keeps a retrospective honest.

A good decision can produce a bad outcome, and a bad decision can be rescued by luck. Judging only by outcome teaches the wrong lesson in both directions.

Ask instead whether the decision was reasonable given what you knew at the time, and whether you could have known more without much effort.

Where the answer is that the decision was sound and the result was poor, there is nothing to change, and saying so protects you from over-correcting.

Where the answer is that ten minutes of checking would have changed it, that is a real finding and it usually recurs.

A worked example

A firm did this properly for the first time, with three people in a room for forty minutes.

The first list had eleven items, most of them small irritations.

Putting a rough cost against each reordered it completely. The expensive ones were a supplier they had kept eighteen months too long, a price that had not moved in two years, and a piece of work they had taken on knowing it was outside what they were good at.

None of those had been in the top five of the original list.

They wrote one sentence for each about what would have to change, and those three sentences became the January priorities.

The irritations, which had dominated the conversation, turned out to cost almost nothing.

Skip the successes

A deliberate omission, because year-end reviews conventionally start with them.

Successes are pleasant to discuss and teach less, partly because it is genuinely difficult to know which decision produced a good result and partly because nobody examines them critically.

Where something worked, the useful question is narrow: can it be repeated deliberately, and do you know why it worked.

If the answer to the second is no, it is luck rather than a method, and treating it as a method is how businesses repeat something that will not work twice.

Ten minutes on that is proportionate. The rest of the time belongs to the other question.

Write what would have to change

The step that turns a reflection into a plan.

For each of the three, write one sentence describing what would have to be different for it not to recur.

Not a resolution, which is a statement of intent, but a change: a rule, a limit, a date, a person, or a process.

Raise prices in January rather than intending to charge more. Decline work outside a stated list rather than being more selective. Review suppliers in December rather than being less tolerant.

A regret with no mechanism attached recurs, and most retrospectives produce exactly that.

Do it with somebody else

A practical note, because doing this alone produces a softer version.

A colleague, a partner, or somebody who works with you will name things you have talked yourself out of, and will notice when an explanation is a rationalisation.

Where there is nobody, writing it down rather than thinking it achieves some of the same, since a written sentence is harder to soften than a thought.

Keep it short and keep it civil: this is about decisions rather than people, and a retrospective that becomes an argument produces nothing and costs goodwill.

Forty minutes is enough, and longer sessions tend to drift into planning, which is a different exercise for a different day.

The counter-case

Retrospectives can do harm.

After a hard year, an exercise focused entirely on regret is demoralising and can produce over-correction: abandoning something that was working, or becoming cautious in ways that cost more than the original mistakes.

There is also a version where the same three items appear every December and nothing changes, which suggests the problem is capacity or willingness rather than awareness.

If that is happening, the honest finding is that these are not going to change, and the useful response is to plan around them rather than to list them again.

Keep it to forty minutes, produce three changes with mechanisms attached, and stop.

The forty minutes

  1. Ask what you would do differently, not what went wrong.
  2. List everything that comes up quickly.
  3. Put a rough cost against each.
  4. Keep the three most expensive.
  5. Check whether each decision was reasonable at the time.
  6. Write the change that would prevent a repeat.
  7. Spend ten minutes on what worked and why.

Step three is the one that reorders the list, and the reordering is usually the finding.

The numbers behind it are covered in a year of analytics, read once.

Frequently asked questions

What is the right question to ask?

What would you do differently, rather than what went wrong. The first assumes hindsight rather than fault, and people answer it honestly, including about themselves.

How many items should the list have?

Three. A long list of regrets produces no change because nothing on it feels achievable and the whole thing reads as failure.

How should I choose them?

By rough cost in money or hours, not by irritation. The thing that annoyed you most is frequently not the thing that cost most, and the reordering is usually the finding.

Where are the answers usually?

A decision deferred for months, a price not raised, a supplier kept too long, work taken on that you knew was wrong, or something promised that could not be delivered.

Should I judge by outcome?

No. A good decision can produce a bad outcome. Ask whether it was reasonable given what you knew, and whether ten minutes of checking would have changed it.

Why skip the successes?

Because it is difficult to know which decision produced a good result, and nobody examines them critically. Ten minutes asking whether it can be repeated deliberately is enough.

West Coast Media Solutions Inc. provides web design, web development, hosting, digital marketing, and business consulting to organisations across Canada, drawing on more than twenty-five years in the field.

Doing a year-end review?

Ask what you would do differently, then put a cost against each. The list reorders completely, and that is the useful part.

Start a Conversation