KPI

A small company needs six numbers: enquiries per month, cost per enquiry, close rate, average order value, cost of acquiring a customer, and the share of enquiries coming from referrals. They fit in one table on one screen. The person watching them should be the one who controls the budget, not the one running the ads. Review monthly, on the same date, even in a weak month. Weak months are the interesting ones.

A manufacturer made four films a year with us and picked views as their KPI. The views grew, the enquiries did not. Once we swapped the measure for the number of people who finished a video and then opened the service page, it turned out two films out of eight were working. The budget did not increase, it moved. You spot a vanity metric with one question: what would I do differently if this doubled tomorrow?

The tooling is modest: GA4, properly tagged campaign links, a spreadsheet, and one question on your enquiry form asking how people heard about you. A paid platform adds nothing until you have volume. From you we need analytics access, numbers from invoicing, and someone who fills in the sales outcomes once a month. Without that last part you have half a picture, and that half lies.

The more common mistake runs the other way: fifteen KPIs in a monthly report. With fifteen you follow none of them properly and the report becomes a ritual nobody reads. Keep three to six. And let us be honest, part of marketing cannot be measured precisely. Brand, word of mouth, somebody remembering you from a trade fair. Anyone promising decimal places on that is either wrong or selling software.

In our tables every KPI has an empty column next to it headed what we did. If nothing appears there for three months, the row goes. It sounds harsh, but it keeps the sheet alive. Try it on your own report this month: next to each number, write one sentence about what you would do if it fell by a third.