Conversion rate optimisation is usually sold as a testing discipline: build two versions of a page, split the traffic, wait for a statistically significant winner. That works for an ecommerce site with fifty thousand visitors a month. For a plumber with eight hundred, a test would need to run for most of a year to tell you anything, and the market would have changed by the time it finished.
Local conversion work has to be done differently. The gains come from removing friction that is obvious once someone looks for it, using evidence rather than experiments.
What is realistic at your traffic level
| What works | What does not | |
|---|---|---|
| Under 2,000 visits a month | Fixing friction on judgement and evidence | A/B testing — results will not be significant |
| 2,000–10,000 | Friction fixes plus testing big changes only | Testing button colours or small wording |
| Over 10,000 | Structured testing becomes worthwhile | Testing without a clear hypothesis |
Where local sites lose enquiries
The losses are remarkably consistent across sectors. A phone number that is not tappable on a mobile. A contact form with eleven fields when three would do. No indication of price, so the visitor leaves to find someone who says. Opening hours that contradict the Google listing. A booking process that requires a phone call during the hours the customer is at work.
None of these needs a test to justify fixing. They need someone to go through the site as a customer would, on a phone, and write down every point at which they would give up.
Measure the right thing
Conversion rate is a ratio, and ratios are easy to improve by accident. Cut your traffic in half by removing a page that attracted the wrong visitors and your conversion rate doubles without a single extra enquiry. What we report is the number of genuine enquiries, where they came from, and whether they became customers — the figures that actually pay for the work.
Where this overlaps with search
Conversion and search are more connected than they look. A page that ranks for the wrong query converts badly because the visitor wanted something else. Fixing that is often a search problem disguised as a conversion one, which is why we look at both together rather than treating the page and the traffic as separate jobs.