Every campaign report contains a number that sounds enormous and means very little. Reach is usually that number. It is the count of people who could conceivably have seen something, and it is the first figure a finance director will take apart.
This is not an argument for abandoning measurement. It is an argument for choosing measures that survive contact with someone whose job is to allocate capital.
Why reach fails the room
Reach fails for three reasons. It counts opportunity rather than attention, so a post scrolled past at speed counts the same as one watched to the end. It is trivially purchasable, which means it can always be increased without anything improving. And it cannot be connected to money without a chain of assumptions, each of which someone will question.
Engagement rate is better but has the same weakness in a different coat. A high rate on a small audience of people who will never buy is a pleasant statistic about a commercially irrelevant group.
What a board is actually asking
Underneath the questions asked in a review meeting there are usually only three. Did this produce sales we would not otherwise have made? What did each of those sales cost us? And if we spend more, does the relationship hold?
Every metric worth reporting answers one of those. Everything else is context, and context belongs in an appendix.
Metrics that hold up
Incremental revenue. Not total revenue with a tracking code attached, but the portion that would not have happened anyway. Establishing it requires a comparison: a holdout region, a period without activity, or a matched audience that saw nothing. It is more work and it is the only figure that answers the first question honestly.
Cost per acquisition, by channel and by creator. Aggregate CPA hides the fact that most programmes have a handful of partners producing acceptable economics and a long tail producing none. The average conceals the decision you need to make.
Contribution margin after fees. A sale at a discount code, minus the commission, minus the cost of goods, minus the agency fee, is a different number from the one on the dashboard. Report the version with everything taken out.
Payback period. How long before a cohort of customers acquired this way covers what it cost to acquire them. This is the metric that decides whether the programme can scale or merely continue.
Repeat rate. Customers arriving through a trusted recommendation often behave differently from paid search traffic. If they come back more, that belongs in the case for the channel. If they come back less, you need to know that before you scale it.
Decide before you launch
The single most common failure is choosing the measure after seeing the outcome. If reach is what looked good, reach becomes the headline. This is not dishonesty so much as drift, and it is why measurement should be fixed in writing before anything goes live.
Write down the primary metric, the target, the observation window, and what you will do if the target is missed. That last item matters most. A metric with no consequence attached is decoration.
Report the uncertainty too
Attribution is imperfect and everyone in the room knows it. A report claiming precision it cannot have invites doubt about the parts that are solid. It is more persuasive to say that tracked sales were one figure, analytics showed another, and the gap is explained partly by discount codes shared beyond their intended audience.
Our piece on attribution without fooling yourself works through where the gaps come from. Stating a range and the reason for it survives scrutiny. A single confident number rarely does.
Where reach still belongs
Reach is a diagnostic, not a result. If a campaign underperformed, reach tells you whether the problem was distribution or persuasion. Few people saw it is a different failure from many people saw it and did nothing, and the fixes are not the same.
Keep it in the report. Keep it out of the headline.