Location targeting on Search: presence only or presence and interest?

Google defaults Search campaigns to 'People in or regularly in your targeted locations', which sounds fine until you open the location report. A local service business in a big city can end up paying for clicks from people who were just passing through, or who are planning a trip somewhere else entirely. 'Presence only' cuts that, but it also cuts people who are planning a visit or researching a service they'll buy later.

My take: for a plumber or a dentist, presence only is almost always right, because the job happens where the customer is. For a hotel or a tour operator, interest can be worth the extra waste, since the search is often the research step. Either way, look at the report by 'Matched location' rather than the campaign total before you change anything. Has anyone seen interest-based targeting pay off for a local business?

Agreed on the plumber and dentist case. I'd also look at the "Matched location" column, filtering to rows where the match is "Location of interest" and the query includes a city name or a region, since that's where planners and travelers show up. If those queries convert at a rate you'd actually take, interest targeting is earning its place. If they're mostly tourist-style searches, it isn't.

The city-name filter has a hole: a query that names a city doesn't mean the searcher is a planner. Someone in town typing "dentist Kadıköy" is just a local using their own district's name. So split the rows by whether the city in the query is inside your service area or outside it. Only the outside group tells you whether interest targeting reaches people who will actually travel to you. For most local services that group will be small, so look at its conversions before you widen anything.

If the customer has to be physically where the service happens, I'd keep presence only and leave it there. Interest targeting earns its place when the buyer will travel to you and the purchase justifies the trip, like a hotel or a clinic people fly in for. For a plumber, most out-of-area searchers can't become customers, so that spend is pure waste.

Iris's split helps, but I'd judge the outside group on cost per conversion against your in-area clicks, not on conversion rate. A small sample can show a healthy rate while the cost per lead is still bad. And if the city in the query is one you truly can't serve, exclude it as a negative location rather than counting on bids to sort it out.

Honestly I'm not sure "most out-of-area searchers can't become customers" holds up. A lot of them are booking for a parent or sibling who lives in your area, or they're about to move there, and those are real leads. Look at the actual queries before you write that spend off as waste.

Fair, I was too quick to write that spend off. What would settle it is whether your booking flow asks for the service address separately from where the person is. If it only captures the searcher's location, a daughter in another city booking a plumber for her mum gets screened out, and you never see that lead in the data. "Outside area" then looks like pure waste when it's really a form problem. Check a few of those queries against the calls and forms that actually came in before you decide.

Agreed on the form, but the Google report can't show this case at all. It only tells you where the searcher was matched, so a daughter in Ankara booking for her mum in Izmir looks like an outside-area click no matter what your form captures. The service address only lives in your CRM or call notes. Have whoever takes the calls tag the ones where the job address is in a different city, then compare those against the search terms. That gives you a real number to act on.

Call tagging is the only way to get that number. One thing that changes the math: have the call takers tag the job address, not just the city. A daughter booking for her mum in Izmir often lives in a suburb that's technically inside your radius while the mum is outside it, and a city-level tag will misread that.

Also check whether the outside-area leads cluster around one or two specific cities. If they do, you can make a targeting call per city instead of a blanket yes or no on interest targeting.

My rule of thumb: keep presence only as the default, and move a city to interest targeting only if its tagged job-address leads cost less per lead than your in-area leads. Per-city samples are small, so one good month can fool you, and comparing cost against your own baseline is harder to kid yourself with than a conversion rate.

The presence or interest setting also lives at campaign level, so acting on a per-city answer usually means a separate campaign for those cities, or excluding them from the main one. Plan for that before you start collecting data, otherwise the answer won't be usable.

Agreed on the baseline, but timing will skew it if you're not careful. Comparing outside-city leads to in-area leads by call date throws the numbers off, because a daughter booking for her mum in Izmir often calls days after the original click. Attribute both groups by click date instead. For that, the CRM has to store the gclid, or your call-tracking click ID, with each lead, so the cost is tied to the click that produced it and not to whatever week the phone rang.

Also, switching a city between presence and interest doesn't rewrite past data. So the per-city baseline has to be collected under the setting you're testing, otherwise you're comparing two different targeting setups and calling it one result.