Find the ad that isn't average

By definition, most ads are average. Write one ad and you are guaranteed to be running it whether it is any good or not. Write two and you have started an experiment: Google shows both, the numbers diverge, and one of them turns out to be quietly better.

The differences are rarely small. Ad copy is the highest variance part of the whole account. A headline that names the actual worry ('Heel pain every morning?') routinely doubles the clickthrough rate of a headline that names the service ('Podiatry services available'). And clickthrough rate feeds Quality Score, which compounds into cheaper clicks across the account.

So the goal is not better ads on average. The goal is to find the one ad that isn't average, retire the rest, and then beat it again next quarter.

Every ad group in your account should have at least two enabled ads at all times. Not sometimes. Always. One ad means no experiment, no comparison, and no way of knowing whether you are running the winner or the dud.

There is a second, less obvious reason. Ads get disapproved, sometimes days after they were approved, and policy sweeps do not send a courtesy text first. An ad group with two enabled ads survives a disapproval. An ad group with one goes completely dark, and nothing about your account dashboard makes that obvious.

Change one thing. If the challenger has a new headline, a new offer and a new landing page, and it wins, you have learned nothing you can reuse. Test the worry against the service name. Test 'book online today' against 'same week appointments'. One variable per round, so the win has a reason you can build on.

Judge on bookings, not clicks. A clever headline can win clicks from people who were never going to book. The scoreboard is cost per booked patient, which means your conversion tracking has to be counting real bookings before any test result means anything.

And let it run. A few hundred impressions per ad is a minimum before the numbers mean much, and on a typical clinic budget that is weeks, not days. Crown a winner only when it leads by a clear margin, not by a nose. I use 20% as the floor: anything closer is noise wearing a trophy.

Modern Google ads are responsive: you supply up to 15 headlines and 4 descriptions, and Google assembles combinations on the fly. Plenty of people concluded that Google now does the testing for you. It does not. It optimises combinations inside each ad. The contest between ads, and between angles, is still yours to run.

Give each responsive ad one coherent angle, expressed through all of its headlines. An ad where every headline circles the same worry beats an ad stuffed with 15 unrelated headlines pointed in 15 directions. Pin sparingly: pinning headlines locks Google's hands, so reserve it for the one thing that must always show, like the clinic name or a compliance line.

Google will also grade your ads with an ad strength label. Treat it as a formatting hint, not a verdict. I have seen 'Average' ads outbook 'Excellent' ones repeatedly. The only label that matters is what a booking costs.

When a test settles, pause the loser. Do not edit it in place: editing an ad resets its history, so you lose the very record that told you it was losing. Write a fresh challenger instead, aimed at the winner, built on what the last round taught you. Two enabled ads, always, with the crown constantly defended.

Where do challenger angles come from? The words your patients use. The worry the parent types at 9pm is different from the service name on your fee schedule, and the ad that wins is almost always the one speaking patient, not clinician. If you are stuck, your search terms report and your front desk are better copywriters than any template.

One boundary before you write anything: ad copy for Australian clinics lives inside AHPRA's advertising rules. No outcome guarantees, no testimonials, no 'best physio in Brisbane'. The good news is that honest, specific copy is also what wins tests. The rules and the results point the same way.