Why location targeting goes wrong
Location targeting looks like the simplest setting in an account: type a town, draw a circle, move on. In practice it fails in a few predictable ways. The first is the presence-or-interest problem. The Google Ads help centre documents that campaigns can target people who are in or regularly in a location, or people who are in it or have shown interest in it. The broader option is useful for a hotel, which wants people planning a trip, and wasteful for a plumber, whose customers must be in the area. Accounts set up quickly often end up on the broader option without anyone choosing it.
The second is the shape. Circles ignore geography. A radius around a shop on one bank of a river includes neighbourhoods on the other bank that are a long drive away by bridge. A radius around a clinic near a county border includes people who use services on their own side. The third is the absence of exclusions: the business does not serve a particular estate, a neighbouring town has its own branch, or a large city on the edge of the radius dominates the audience and the spend.
The fourth is scale. A local audience is small. On search, that means few impressions and a quickly exhausted market. On social and display, it means the same people see the ad over and over, and frequency (FrequencyAverage times a person saw an ad.) climbs until the ad becomes wallpaper. The fixes below address each of these in turn. Campaign structure and bidding in general belong to the platform sections; this guide covers the location decisions only.
Targeting options by platform
| Platform | Location options documented in the help centre | Local-specific features | Watch for |
|---|---|---|---|
| Google search | Countries, regions, cities, postcodes, radius around an address; presence or presence-or-interest; exclusions | Location assets linked to the business profile, call assets, location-based reports | The presence-or-interest option; broad match reaching unrelated places |
| Google display and video | Same geographic options as search, combined with audiences and content targeting | Location assets in some formats; local creative variants | Small audiences and high frequency; placements far from local intent |
| Performance Max and similar automated types | Location settings at campaign level; store goals where eligible | Store visit and local action goals for eligible advertisers | Automation spending where the conversion is easiest, not where the business wants it |
| Meta (Facebook and Instagram) | Countries, regions, cities, postcodes and radius around an address or pin; exclusions | Local creative, store traffic objectives where available, messaging and call buttons | Options about who counts as being in the location; small audiences fatiguing fast |
| Other social and display networks | Usually country, region, city and postcode; radius on some | Varies | Coarse targeting that cannot follow a real catchment |
The specifics change as platforms update their products, so treat the help centres as the source of truth. The display targeting options guide covers how Google's geographic targeting combines with audiences and content, and the Meta custom audiences guide covers building audiences from customer lists and site visitors, which can be combined with a location fence for remarketing to local prospects. Housing ads are an exception, where postcode and audience options can be limited, and the real estate marketing guide explains how agencies plan location targeting within those limits.
Radius, exclusions and presence vs interest
Start with presence. For any business whose customers must be physically in the area, set the Google Ads location option to people in or regularly in the targeted locations, and check it in every campaign. Then look at the location report after a few weeks, which shows where the people who saw and clicked the ads were. If clicks are arriving from far outside the area, either the option is wrong or the targeted locations are too broad.
Draw the area from data, not from a compass. Plot where existing customers live or where jobs were done, and look at drive times. The useful area is usually irregular: postcodes along a main road, a few towns that share a shopping centre, a district that has no competitor of its own. Build it from postcodes or several small radii rather than one large circle. Then add exclusions for the places the business does not want: areas served by another branch, areas with access problems, areas where the business does not hold a licence or does not go. The same job records decide which areas deserve their own page on the website, as the guide to service area pages explains.
Adjust bids or budgets by location where the data supports it. Areas closer to the shop or with more valuable customers can carry higher bids on search; the location report shows cost per click (CPCAverage cost of each click.) and conversions by area, and the CPC calculator helps compare what each area is worth paying. On social and display, the opposite constraint applies: a radius too tight for the budget concentrates impressions on a tiny audience. Widen until frequency stays sensible and let the creative do the local work.
Location assets and local creative
On search, location assets attach the business address, distance and a directions link to the ad, drawn from the linked business profile. The help centre documents linking the profile to the Ads account to enable them, and affiliate location assets for brands sold through retail chains. Call assets add a phone number or a call button. Both make a local ad look local, which matters when the person is choosing between a nearby shop and an online retailer. The ad assets guide covers setting them up and how they appear.
Creative should prove locality, not just claim it. Name the area in headlines and copy where it is natural: the town, the district, the landmark people use for directions. The local keyword research tutorial shows which place names customers actually search with. Use photos of the actual shop, the actual team and the actual vans, not stock images of generic interiors. Mention local specifics that a national competitor cannot: same-day visits within the area, a branch on the high street, parking behind the shop. On social, short video of the place itself tends to outperform polished brand creative because it looks like something a neighbour would post.
For several locations, use one campaign or ad set per location, or location-specific creative within a shared structure, so that each area sees its own address and its own offer. An ad for the northern branch shown to people near the southern one sends them to the wrong door, or to a competitor.
Store visit and call goals
A local campaign judged on website conversions alone undercounts most of what it does. The person who saw the ad and drove to the shop, or saw the number and called, often leaves no trace on the website. Measure the local actions directly.
Calls. Track calls from call assets and from the website number with a forwarding number the ad platform provides, or with a call-tracking service. Set a minimum call length that counts as a conversion so that wrong numbers and hang-ups do not inflate results. How to attribute those calls by source, judge their quality and join them to the CRM is covered in the guide to measuring phone calls and chat enquiries.
Directions and visits. The Google Ads help centre documents store visit conversions for advertisers who meet its eligibility requirements, which depend on factors such as having verified locations and enough visit data. Where eligible, they give a modelled estimate of visits after ad interactions. Where not, use direction requests from location assets and from the profile, and compare them across periods.
Offline signals. Offer codes, a question at the till or on the booking form, and comparisons of sales in weeks with and without spend all help. None is precise; together they show direction.
On social and display, reach (ReachUnique people who saw an ad.) and cost per thousand impressions (CPMCost per 1,000 impressions.) describe delivery, not results. The reach and frequency calculator helps plan how many people in the area a budget can reach at a sensible frequency, but the judgement should rest on calls, visits and bookings.
A dental practice tightens its fence
Riverside Dental, a single practice in a market town, ran a search campaign targeting a 25-mile radius with the presence-or-interest option. The location report showed that 38% of clicks came from a city 30 miles away, where people searched for dentists in the town while planning a visit. The practice switched to presence only, replaced the circle with nine postcodes covering the town and the villages that used it, excluded the city, and added location and call assets. Monthly spend stayed at £900. Clicks fell from 410 to 280, but calls of over a minute rose from 31 to 54 and new patient bookings from 19 to 33.
Budgets across locations and seasons
A budget spread evenly across areas and months will be wrong in most of them. Allocate by demand and capacity. Areas with more searches, more valuable customers or less competition deserve more; areas where the business is already full, or which another branch serves, deserve less. Impression share (Impression ShareShare of eligible impressions you received.) lost to budget in a strong area is a signal to move money towards it; impression share lost to rank says the ad or the bid needs work instead. When the question is how a national brand plan should split its weight between regions rather than between a business's own areas, the tutorial on weighting a media plan by region works it out from sales, category and distribution data.
Seasonality is usually stronger locally than nationally, because local demand follows weather, school terms, tourist seasons and local events. Plan the year as a media plan, with spend weighted towards the weeks of highest demand, and check search trends for the area rather than assuming. For social and display, raise frequency for short periods before a peak rather than running a thin presence all year.
Keep lead units and geo-targeted campaigns in one plan. A service business running Local Services Ads and a search campaign side by side should judge them together on cost per booked job, and move money between them as the results suggest. The local advertising hub sets out the overall budget logic.
To prove that local media moved footfall rather than only reporting modelled visits, the measurement plan for store visits sets out proxies, regional sales matching and holdout regions.
For a Google-only build, the Google Ads setup for store visits combines linked locations, per-store catchments, local-action goals and opening-hour schedules into one recommended configuration.
The store visits objective hub lists the channels that can drive footfall and the KPIs they share, and the paid search channel page for store visits sets per-store catchment, query and budget tactics for search.
Service businesses choosing between a location-fenced search campaign and Google's pay-per-lead unit can use the comparison of Local Services Ads and search ads to decide which suits their trade.
Common mistakes
- Leaving the presence-or-interest option on. The single most common source of out-of-area clicks for local businesses.
- Drawing one big circle. Catchments follow roads, rivers and habits; build the area from postcodes and data.
- No exclusions. Areas served by another branch, or not served at all, keep absorbing spend.
- Generic creative. An ad that could be from any business in any town gives the local business no advantage.
- Judging local ads on website conversions alone. Calls, directions and visits are most of the result.
- A radius too small for the budget on social. The same few thousand people see the ad until they stop noticing it.
- Even budgets across areas and months. Demand is uneven; the budget should be too.
- Never reading the location report. It is the only way to see whether the fence is where it was meant to be.
