How to Run a Weekly Growth Review That Changes Decisions
The cadence, the one-page input, and the room for a weekly growth review that ends in decisions instead of a status update nobody acts on.
By The Spend Report Editorial Team. Published June 25, 2026. · 5 min read
On this page
Most weekly growth meetings are reporting theater. Someone walks the room through last week's numbers, everyone nods, a few things get noted, and nothing about how the business runs is different on the way out than it was on the way in. The numbers were reviewed. No decision was made. That is not a growth review, it is a standup with charts.
A real weekly growth review exists to do one thing: convert the last seven days of data into a small number of decisions that change what happens in the next seven. This is how to run one that earns its hour.
The test for whether it is working
Before the format, the standard. A growth review is working if you can answer one question at the end: what are we doing differently this week because of what we just looked at. If the answer is nothing, the meeting failed, no matter how clean the dashboard was. Hold every choice below against that test. Anything that does not move you toward a decision is ceremony, and ceremony is what makes these meetings die.
The one-page input
The review runs off a single page, prepared before the meeting, the same shape every week. Not a live dashboard you explore together, not a 40-tab spreadsheet. One page, sent in advance, that everyone has read before they walk in. The discipline of fitting it on a page is the discipline of knowing what actually matters.
The page holds four things and nothing else:
- The number you run the business on, with its trend. Blended CAC or MER, this week against the last several, so the line is visible, not a single snapshot. If you have not settled which number governs, blended CAC or MER is the prerequisite read, because a review without a primary number drifts.
- The two or three drivers underneath it. Spend, new customers, the efficiency by your main channels. Enough to explain why the top number moved, not the full account.
- What changed last week, and the early read. The tests you launched, the creative you shipped, the budget you moved, and the first signal on each. This is the part most reports omit and the part that connects action to outcome.
- The open decisions. The two or three calls that need to be made this week, stated as questions with an owner attached.
Everything that does not fit on that page lives in a system someone can open if a question demands it. It does not get walked through in the room.
Who is in the room
A decision meeting needs the people who can make decisions and as few others as possible. The room is whoever owns the number and whoever owns the levers under it. For most brands at $1M to $10M a month that is the founder or head of growth, the person who owns paid, and whoever owns creative or site, depending on the week's open decisions.
Two rules keep it sharp. Every person in the room can either decide something or owns something being decided. And if your agency runs paid, they are in this meeting or feeding the one page into it, because a growth review that does not include the people executing the spend is reviewing a story, not the work. Reading an agency performance report is the companion skill for keeping that input honest.
The agenda that produces decisions
Thirty to forty-five minutes, same order every week, because a predictable structure is what lets people prepare and what stops the meeting wandering.
- Five minutes: the top number and trend. Is it where we expected, and if not, why. No narration of figures everyone already read. Straight to the gap between expected and actual.
- Ten minutes: what we changed and what it did. Walk the tests and moves from last week and their early read. Name what worked, kill what clearly did not, and resist the urge to relitigate decisions the data has not weighed in on yet.
- Fifteen minutes: the open decisions. The core of the meeting. Take each open question, make the call, assign the owner, set the check date. A decision without an owner and a date is a wish.
- Five minutes: next week's tests. What are we deliberately trying, so next week's review has something real to read.
Notice what is absent. No open-ended exploration, no pulling new reports live, no status updates that could have been an email. Exploration is valuable and belongs in a different, less frequent meeting. The weekly cadence is for decisions.
The failure modes to watch
Three patterns turn a good review back into theater, and they creep in slowly.
- The number creep. The page grows. One more metric, then another, until it is a dashboard again and nobody can say what matters. Prune it back to the four things every quarter.
- The decision dodge. Open questions get discussed and then deferred, week after week, because deciding is uncomfortable. A review that defers the same call three weeks running has a courage problem, not a data problem.
- The attendance creep. The room fills with people who are informed but cannot decide. Every added observer lowers the odds a hard call gets made. Keep it small on purpose.
How to use this
Build the one page first, this week, in the shape above, and send it before the meeting. Run the fixed agenda for a month before you judge it. The goal is not a better-looking review, it is a review that ends in a short list of owned decisions with dates.
Make sure the top of that page is the right number by reading blended CAC or MER first. If an agency runs your spend, reading their performance report and the first 90 days playbook keep their input feeding your decisions instead of replacing them. And if Amazon is part of the mix, the TACoS calculator gives you a clean efficiency number to put on the page.