Google Ads location targeting: stop paying for out-of-area leads.

For a contractor that only serves jobs inside a defined territory, start by testing Google Ads’ “Presence” option: people likely to be in or regularly in the locations you target. Build the map from real service limits, add exclusions for places you cannot serve, and review matched locations alongside the actual job address. Do not expect a radius or a location setting to be perfectly accurate.

A roofing, HVAC or plumbing campaign can have sensible keywords and strong ads yet still waste money because its map does not match the operation. A 30-mile radius may cross a state line, bridge, toll road, franchise boundary or mountain route. A city target may include neighbourhoods the team cannot reach within its promised response time. A lead can sound local on the phone but provide a job address outside the territory.

Location targeting is therefore not merely an advertising setting. It is a decision about which jobs the company can serve reliably and profitably.

Why Google Ads can show beyond the area you expected

Google’s default advanced location option for Search includes people likely to be in or regularly in the targeted area and people who have shown interest in it. Google says that interest may be inferred from search terms, recent presence or content viewed about the location.

That is why a person outside Dallas who searches for “roofer Dallas” may be eligible for a campaign targeting Dallas under the default option. The search may be entirely legitimate: a landlord, adult child or property manager could be arranging work for a building there. It could also be irrelevant to a contractor that only wants nearby owner-occupiers.

Google offers a narrower “Presence” option for people likely to be located or regularly located in the targeted places. For a service business that travels to the property, this is usually the clearer starting test. It is still not a hard fence. Google explicitly says location targeting uses several signals and that 100% accuracy is not guaranteed.

Choose Presence or Presence or Interest deliberately

Presence makes sense when the service can only be delivered at a property inside the territory and out-of-area interest rarely becomes valid work. Emergency plumbing, same-day HVAC repair and tightly licensed work commonly fit this pattern.

Presence or Interest may be defensible when people outside the area often arrange work for properties inside it. Examples include property managers, owners of second homes, family members coordinating repairs, or a commercial buyer searching from a head office. It can also increase reach when a campaign targets a destination or location-dependent service.

Do not decide from click volume. Run a controlled comparison only when the business can label the property address, lead quality, booking and sale. If broader targeting creates more conversions but most are unserviceable, the conversion count is misleading.

Build the map from operations, not from a circle

Before opening Google Ads, divide the territory into three lists:

• Core areas: places the team can serve consistently, with workable drive time, licensing and job economics.
• Conditional areas: locations that work only for higher-value jobs, certain days, specific crews or planned appointments.
• Excluded areas: places outside the licence, franchise, staffing, response-time or profitability boundary.

A roofing company may travel farther for a full replacement than for a minor repair. An HVAC installer may accept replacement enquiries across a wider region while keeping no-cool repair inside a tighter zone. A plumbing business may offer scheduled repiping farther away but restrict emergency call-outs to the nearest postcodes.

Those are different campaigns because they have different geographic economics. One giant radius forces Google to treat them as if the service limits were identical.

Radius, city, county, ZIP code or postcode?

A radius is quick to set up, but distance “as the crow flies” does not equal travel time. Water, traffic, bridges and road networks can make the near edge unworkable while a farther point along a motorway remains viable.

Named cities, counties, regions, ZIP codes and postcodes can better reflect operational boundaries where Google supports those target types. They are not always exact representations of dispatch reality either. Postal areas may be irregular, large or unavailable at the granularity you expect.

Use the most stable target units that match the business, then apply exclusions where the shape spills into unserviceable areas. Keep a written source of truth outside Google Ads—a service-area sheet owned by operations, not a memory held by the agency.

Do not make the map so fragmented that it cannot gather usable data. A small campaign split across dozens of tiny locations may produce thin reporting and needless management. Separate areas when the service, budget, staffing or value genuinely differs.

Exclusions are useful, but they are not magic walls

Google says its default exclusion behavior removes people likely to be located in the excluded area. Someone outside an excluded area may still see the ad. Because location is inferred, exclusions improve control without guaranteeing perfection.

Exclude obvious non-service territories: neighbouring states or provinces, counties beyond a franchise boundary, cities across an expensive crossing, and locations where licensing prevents work. Avoid building a giant negative map without checking whether it overlaps wanted areas.

If one town generates poor results, diagnose the reason before excluding it. The problem may be a search-intent mismatch, slow response, a service the branch does not perform or a landing page that overpromises coverage. Geographic exclusion should solve a geographic problem.

Targeted locations and matched locations are different reports

Google Ads provides two useful geographic views. “Targeted locations” reports performance for the locations configured in the campaign. “Matched locations” shows the locations where ads appeared, which can reflect a person’s physical location or a location of interest.

That difference answers two separate questions:

• Targeted locations: How did the areas we chose perform?
• Matched locations: Which places did Google associate with the people who actually saw the ads?

Google notes that reports may include “Unspecified” areas when it cannot determine a location from the IP address or query, or when a Maps search covers several areas. Data below country level may not add up to campaign totals. Treat the report as evidence, not a perfect GPS log.

For each lead, compare the matched location with the service address captured by the office. The property address is what determines whether a contractor can do the work. The caller’s detected location is only one clue.

Why location keywords do not replace location settings

Adding “Austin,” “Toronto” or “Birmingham” to a keyword can clarify intent, but it does not restrict the searcher’s location by itself. Someone anywhere can type a city name. Likewise, a person standing inside the service area may search “furnace repair near me” without naming the city.

Use query targeting and geographic targeting together. Review the search terms report for place names, “near me” language and services that imply a different territory. Our guide to contractor search terms and negative keywords explains how to classify those queries without blocking good demand too quickly.

The destination page should also state the service area honestly. A location setting cannot repair a page that implies nationwide coverage. Use our guide to choosing a homepage, service page or dedicated landing page for each campaign.

Measure geography by booked-job economics

Click-through rate and cost per lead do not reveal whether a territory is valuable. Track at least the service address, requested job, qualified status, booking, estimated or invoiced value where available, and the reason for disqualification.

A distant postcode may have a high cost per lead but produce profitable installations. A nearby suburb may send cheap calls for work the company does not perform. Bidding and coverage decisions should reflect suitable jobs, not the cheapest form submission.

Google Ads supports conversion value rules that can adjust reported values based on conditions including location. That can be useful for mature accounts using value-based bidding, but it should come after accurate conversion values and lead outcomes—not as a substitute for them.

For phone-heavy campaigns, connect geography to the full call outcome. The contractor call-tracking scorecard separates rings, connected calls, qualified opportunities, bookings and won work.

Trade-specific examples

US, Canada and UK checks

United States: map state and municipal licensing, franchise territories, tolls and ZIP-level job outcomes. Do not assume a metro name describes one workable dispatch area.

Canada: province, weather, travel distance and bilingual service capability may alter the map. Large postal regions and seasonal routes deserve manual lead review.

United Kingdom: use postcodes and realistic call-out zones, accounting for congestion, river crossings and clean-air or charging zones where relevant. Keep boiler, drainage and emergency coverage aligned with actual engineers.

A 20-minute agency audit

• Open every active campaign’s advanced location option.
• Record whether it uses Presence or Presence or Interest.
• Compare targets with the current operational service-area sheet.
• Review exclusions for gaps and accidental overlap.
• Open matched locations and inspect spend, conversions and unspecified traffic.
• Sample recent leads and compare detected location with the property address.
• Separate urgent, planned and high-value services where their travel limits differ.
• Confirm landing pages and call scripts state and qualify the territory.
• Report qualified leads, bookings and revenue by area—not clicks alone.

The outcome should be a map the operations team recognises, a clear reason for the advanced setting, and a monthly process for learning where profitable work actually comes from.

Frequently asked questions