"Marketing is working" is a feeling. Cost per qualified lead by source is a fact. The gap between them is where most small-business budgets leak. This is a 30-day plan to close it — no analyst required.
Week 1 — tag everything
Every lead that arrives needs to carry the source it came from: the campaign, the ad, the referrer. If you cannot say where a lead came from, you cannot say what it cost. Route all enquiry forms through a system that stamps the source automatically (UTM tags do most of this). Our lead tracking writes the source onto every enquiry the moment it lands.
Week 2 — watch cost per qualified lead, not cost per lead
Cost per lead hides the truth. A £2 lead that never buys is dearer than a £20 lead that always converts. Qualify the lead (a score, a status, a "booked" flag) and divide spend by qualified leads only. The CPQL guide walks the formula; the calculator does it for you.
Week 3 — rank your channels honestly
With two weeks of tagged, qualified data, rank every source by cost per qualified lead and by close rate. The channel with the most clicks is usually not the channel with the best leads. Attribution models (first-touch, last-touch, multi-touch) each tell a slice of the story — the attribution explainer covers when to trust which. Start with last-touch; add multi-touch when you have volume.
Week 4 — move the money
Cut or cap the channel with the worst cost per qualified lead. Shift that budget to the best one. Do not "test" forever — a 30-day window is enough to act, and you can always reverse it. The point is to make next month's spend based on this month's evidence, not last year's habit.
The honest limit
Attribution is not truth; it is the best available inference. A lead who found you by word of mouth but clicked a retargeting ad will be miscredited. That is fine. You are not auditing a tax return — you are deciding where next month's budget goes, and "roughly right by source" beats "confidently wrong by gut". ROI & attribution shows the number next to each channel so the decision is obvious.
Why 30 days is enough
You are not proving causality to a journal. You are building a habit: tag, qualify, rank, move. After one cycle it is just what you do. The teams that win at this are not the ones with the biggest stack — they are the ones who look at the number every week and act on it. Start the trial, tag this month's leads, and by this time next month you will know which campaign actually pays.