A good review does two things at once: it shows the work, and it shows the thinking behind the work. Most decks only manage the first. Here is the structure we use at Proper to do both without doubling the length.

Introduction

Every review starts from the same place: the people in the room did not watch the work happen. They arrive with their own context, their own deadlines, and about four minutes of patience. The job of the first slide is to give them a reason to spend the next twenty.

So we open with the decision rather than the process. What are we asking for, what changes if we get it, and what happens if we do nothing. Everything after that slide exists to support the answer, and anything that does not gets cut before the meeting rather than skipped during it. You can read more about how we structure research before a review.

Sequence matters more than polish. When a reviewer can predict what comes next they stop bracing for surprises and start giving you real feedback, which is the only thing the meeting was ever for.

Write the objections down first. Ask the two people most likely to disagree what they would push back on, put those answers in the deck, and credit them by name when you present. Disagreement handled in advance reads as rigour; disagreement handled live reads as a gap.

Keep one idea per slide and one sentence per idea. If a slide needs a paragraph it is really two slides, and the second one is usually the interesting half. Long appendices are fine — nobody minds page forty as long as pages one through ten were honest.

Rehearse the transitions rather than the words. The sentences come out differently every time, but the order does not, and the order is what people repeat a week later to someone who was not in the room.

Software and tools

We build decks in whatever the team already has open. Tooling arguments cost more hours than they save, and no reviewer has ever approved a budget because the typography was set in the right application.

What does matter is a shared source for the numbers. Every figure links back to the dashboard that produced it, so when someone asks where sixty-two per cent came from the answer takes four seconds instead of a follow-up email.

Other resources

If you are putting a review together for the first time, three habits carry most of the weight:

1. Write the recommendation before you write anything else.
2. Show one artefact per claim — a clip, a chart or a quote — and never more than one.
3. End with the decision you need and the date you need it by.

None of this is unique to design reviews. The same shape works for engineering proposals, pricing changes and hiring plans, because the constraint is identical: attention is short and trust is earned in the first minute.

Start with the ask, defend it with the smallest possible amount of evidence, and leave the room with a decision. Everything else is decoration.