Location targeting for a local clinic: presence only or presence or interest?

For a local service account, Google Ads defaults to "presence or interest", so people who've searched for or read about your city can see your ads even when they're nowhere near it. For a dental clinic or a law firm in one city, that often means clicks from people three states away who typed the city name for a news story or a trip. Switching to presence only cuts that, but it can also cost you volume in edge suburbs where people actually commute in. My lean is to start with presence only for anything people need in person, then watch the search terms report for a few weeks before widening. Has anyone seen "interest" clicks convert at all for appointment-based businesses, or is it mostly wasted spend?

Honestly it depends on how far patients will really travel for the service. A general dental practice mostly serves people within a 20 to 30 minute drive, so clicks from outside that area are usually noise. A clinic known for implants or cosmetic work can pull from a much wider radius, and those people often search the city name while they're comparing options. Rather than going by where your clicks come from, look at where your converted patients actually live. Segmenting conversions by user location in the Geographic report will show whether the "interest" clicks ever turn into bookings.

Using the conversion address makes sense, but there's a catch. The Geographic report's user location is only as good as the signal behind it, and for a clinic a lot of bookings come in by phone call or form. Call tracking and form fills don't always carry location with them, so you end up segmenting only the conversions that happen to have it. With a couple of hundred bookings a month that's probably fine. With ten a month, the sample is too thin to separate "interest" traffic from noise, so I'd check that your location data covers most of your conversions before trusting the split.

I'd fix this upstream of the report. If the booking form asks for a postcode, you can send it back with the conversion through enhanced conversions or an offline import keyed to the click ID, so location travels with the booking. Phone bookings are harder because caller location data is patchy. Asking for the postcode on the call, or logging it in the CRM at intake, covers more of them. Otherwise you're mostly segmenting the easy conversions and reading a pattern into them.

Agreed on the postcode. Enhanced conversions match on hashed email or phone and can pick up the address fields too, but only if the form captures them in a way the tag can actually read. So run a test booking end to end and confirm the conversion shows up before you trust any location split.

Depends on where the booking actually happens. If it runs through a third-party widget or a calendar on another domain, the tag often can't read the fields, so your own test booking passes while real ones still lose location. Watch the match rate for a week instead of trusting one test.