The Weekly Marketing Review: Turn Dashboards Into Decisions

A weekly marketing review turns dashboards into decisions when it follows a fixed ritual: the same 45 minutes every week, one scorecard built on a handful of metrics tied to money, a named owner for every metric, and a decision log that gets reviewed at the following meeting. Skip that structure and dashboards get admired instead of acted on, and the numbers you paid to collect change nothing about how you spend.

Why dashboards alone change nothing

Most marketing teams do not have a data problem. They have a decision problem. The dashboards exist, the connectors run on schedule, GA4 and the ad platforms are wired up, and yet budgets roll over untouched. A dashboard is a passive object. It answers questions only when a human asks them, and only in a setting where the answer is expected to produce an action.

The weekly marketing review is that setting. It is not a status update for stakeholders and it is not reporting theater where someone walks executives through pretty charts. It is the recurring hour where a team converts last week's numbers into next week's moves. Everything about the ritual, from the scorecard to the agenda to the log, exists to force that conversion instead of leaving it to chance.

The scorecard: fewer numbers, tied to revenue

The review runs on one prepared scorecard, not on live dashboard spelunking. Keep it to roughly eight to twelve metrics that connect to money: blended spend, backend revenue and its marketing-attributed share, a blended efficiency metric such as MER or blended CAC, and two or three leading indicators for each priority channel. Every metric carries an owner and a target, and shows its trend against prior weeks rather than a lonely raw number that means nothing without context.

Everything else belongs in an appendix, including platform-reported ROAS by campaign. Platform numbers are genuinely useful for in-channel tactics, but the scorecard should stand on backend truth, because summed platform conversions routinely exceed what actually landed in your billing system. If the headline scorecard is built on numbers each ad network grades for itself, the meeting inherits every incentive those networks have to flatter their own performance.

A metric without an owner is a fact nobody is accountable for. Assign each line to someone who can explain the movement and propose an action, and pair it with a target so the room sees at a glance whether the number is a problem or just weather. Green metrics that hit target get zero airtime; the review exists to spend attention on the exceptions.

The 45-minute agenda

1. Five minutes: scorecard scan. Everyone reads the pre-circulated scorecard. Metrics that are green and on target get no discussion.
2. Ten minutes: last week's decisions. Walk the decision log out loud. Did we do what we said we would? What actually happened, and did it match what we expected?
3. Twenty minutes: the two or three red metrics. The owner of each red metric presents a diagnosis and a proposed action, not a narration of the chart everyone can already see.
4. Ten minutes: decisions and owners. Every discussion ends as a decision, an experiment, or an explicit watch-one-more-week, written down with a name and a date attached.

Two rules keep the meeting honest. First, no live debugging: when a number looks wrong, log it as a tracking task and fix it offline rather than burning the room's time chasing it. Second, no re-litigating strategy every week. Channel-mix architecture and budget splits belong to a monthly or quarterly review, where MMM outputs and incrementality results carry far more weight than seven days of noisy fluctuation.

The decision log is the actual product

The most valuable artifact the ritual produces is not the dashboard and not the scorecard. It is a running decision log with four columns: date, decision, owner, and expected result. Reviewing that log at the top of every meeting creates the feedback loop most teams never build. You find out whether last week's calls were right, and your forecasting sharpens because every prediction gets checked against reality within seven days instead of being quietly forgotten. Over a quarter the log becomes an honest record of judgment, surfacing which channels beat their forecasts and which bets keep disappointing.

Making the ritual stick

Same day, same time, every week, and the meeting happens even when the calendar looks ugly, especially during peak periods like Black Friday when the temptation to skip process is strongest and the cost of a bad call is highest. Circulate the scorecard the evening before so the scan is genuine reading, not first impressions. Cap attendance at the people who own metrics plus one decision-maker who can approve budget moves on the spot.

If the meeting regularly ends with no decisions, the scorecard is too big or too soft. Shrink it until every number on it can force an action from someone in the room. A tight review that produces three real decisions beats a sprawling one that produces admiration.

Where a unified measurement layer helps

The ritual only works if the scorecard is trustworthy, and that is where most teams struggle: attribution lives in one tool, MMM in another, lift tests in a spreadsheet, and none of them agree. Admira gives the weekly review a single reconciled scorecard, with multi-touch attribution, marketing mix modeling, and incrementality results unified in one place and checked against backend revenue, live in about two weeks. With one honest source feeding the meeting, the room can spend its 45 minutes deciding instead of arguing about whose number is right. To give your weekly review one scorecard everyone believes, book a demo.

FAQ

Which metrics belong on the weekly marketing review scorecard?

Keep it to eight to twelve numbers tied to money: blended spend, backend revenue and its marketing-attributed share, blended efficiency such as MER or blended CAC, and two or three leading indicators per priority channel. A simple test decides the rest: if a metric could double without anyone changing behavior, it is context, not a scorecard line.

Who should run the weekly marketing review?

The most senior person accountable for marketing performance owns the meeting, but the individual metric owners do the talking. A review where one person narrates every chart while everyone else listens is a status meeting wearing a disguise. The chair keeps time, forces each discussion to end in a decision, and makes sure the log is written before anyone leaves.

What if our tools disagree on the numbers?

Pick one source of truth for the scorecard, ideally backend revenue viewed through a single consistent attribution lens, and note the known gaps beside it. When Meta, Google Ads, and GA4 disagree, do not debate them live. Log the discrepancy as a tracking task, assign an owner, and move on. Reconciling platform numbers is real work, but it belongs offline, not in the 45 minutes reserved for decisions.

Should the marketing review be weekly or monthly?

Both, with different jobs. The weekly review is tactical: pacing, creative fatigue, anomalies, and fast course corrections. The monthly or quarterly review is strategic: budget mix, channel architecture, MMM readouts, and incrementality results that are too slow to move week to week. Mixing the two altitudes in one meeting ruins both.

How do we keep the weekly review from becoming a status meeting?

End every discussion in a written decision, an experiment, or an explicit watch-one-more-week, each with a named owner and a date. Ban live debugging and strategy re-litigation. If the meeting regularly closes with no decisions, your scorecard is too big or too vague, so shrink it until every remaining number can force an action from someone in the room.