Which marketing metrics flatter and which tell the truth
Learn to pair every flattering marketing metric with the number that tests it so your monthly report stops disagreeing with your bank balance.
Rising numbers and a flat bank balance
The monthly report arrives and everything in it is up. Impressions up. Reach up. Followers up. Engagement rate holding. Then you look at collections for the month and nothing has moved. Nobody in this situation is lying to you. They are reporting the numbers the platforms put in front of them.
Marketing metrics divide into two groups. Some measure activity and some measure outcome. Activity metrics are easier to collect and easier to improve and they are not worthless. They are just not evidence. The fix is not to ban them. It is to force each one to sit beside the number that tests it.
What honest reporting costs
This change costs no licence fee and that is the trap. It costs something harder to pay. Someone has to lose an argument and a channel someone defends will start to look weak. Expect that before you begin and the exercise survives contact with the team.
The practical cost is joining data. Spend sits in ad platforms. Revenue sits in a gateway or an accounting system. Qualified enquiries sit in a CRM or a WhatsApp inbox. Somebody has to pull three sources into one view each month and that is a few hours of work that nobody currently does. Your agency's standard report may not include it either so a custom monthly view can carry its own fee.
When activity metrics are the right answer
If you are three months into a brand launch do not demand a cost per acquisition from your agency. There is nothing yet to divide. Ask instead for distribution and for evidence that the right people saw it. Insisting on efficiency metrics during a launch produces exactly one behaviour. The team shifts budget to retargeting and brand search because those convert and your launch becomes a harvesting exercise with nothing planted.
Long sales cycles deserve the same patience. An industrial supplier with a nine-month cycle cannot be judged monthly on closed revenue. Leading indicators are legitimate there. Qualified enquiries. Site visits from target company profiles. Quotation requests. The discipline is to name which leading indicator you believe predicts revenue and then check the belief once a year.
Everyone else should be running on outcome metrics and most are not.
Pairing each metric with its test
Work through your report and write the testing number beside each line. Impressions are tested by qualified enquiries. Follower growth is tested by repeat purchase rate. Click-through rate is tested by cost per qualified lead. Sessions are tested by conversion rate on the pages those sessions land on. Return on ad spend reported inside a platform is tested by blended cost of acquisition across all spend. Email open rate is tested by revenue per send. Keyword rankings are tested by non-brand organic sessions reaching commercial pages.
Then give every surviving line an owner and a decision. If no decision in the business changes when a number moves then the number is in the report for comfort. Delete it. A report that loses half its rows usually gains all of its influence.
Judging the report itself
The metric that matters here is whether anything changed. Count the decisions taken in the last quarter that trace to the report. Budget moved. A page rebuilt. A channel paused. Zero means the report is a ritual.
The metric that flatters is the report's own thoroughness. Page count. Chart count. The number of channels covered. A forty-page monthly deck signals effort and usually signals that nobody was willing to say which three numbers mattered.
The common mistake
The expensive error is letting the tool choose the metric. Every platform surfaces the numbers it can reliably improve and those are rarely the numbers that pay you. Optimise to the platform's convenient metric and it will deliver. Cheap reach. Cheap clicks. Cheap video views from audiences that will never buy. You will have bought exactly what you asked for.
Averages do the second most damage. Average order value hides the difference between many small orders and a few large ones. Average session duration hides the difference between careful readers and people stuck on a broken page. Look at distributions and at segments before you move money. Month on month comparison in a seasonal business misleads the same way so compare with the same month last year wherever you have the history.
Revenue is not what you keep
One last correction worth making. Judging marketing on revenue overstates performance because revenue is not margin. GST collected from a customer is not yours to count since it is collected and remitted. Shipping costs and payment gateway charges and returns all sit between a reported sale and money you keep. Marketing judged on contribution after those deductions produces different channel rankings to marketing judged on revenue and the difference is sharpest in low-margin categories. How these items are treated in your accounts depends on your registration and your structure so a qualified professional should confirm the right basis for your business.
This guide explains how something works. It is not legal, tax or financial advice, and rules change. Confirm anything regulatory with a qualified professional before you act on it.