Geofencing ads let you show ads to people based on where their phone has been. You draw a virtual boundary around a place, such as your store, a competitor's showroom, a trade show hall or a stadium, and anyone whose device enters that boundary becomes part of an audience you can reach on apps and websites, sometimes for weeks afterwards.
It sounds like the most precise targeting in advertising, and sometimes it is. It is also one of the easiest channels to waste money on, because the reporting looks impressive and the underlying data is noisier than most vendors admit. This guide covers how geofencing actually works, what it costs, where it earns its keep, and how to judge whether a campaign paid for itself.
What Geofencing Ads Are (and What They Are Not)
People use "geofencing" loosely, so it helps to separate three different things that often get sold under the same name:
- Geofencing (real-time). A device is inside the fence right now, and the ad is served while it is there. Useful for "you're nearby, come in" messages.
- Geo-retargeting or "geo-framing". You define a fence and a past date range. Devices seen inside the fence during that window go into an audience, and you serve them ads later, wherever they are. Event retargeting and conquesting competitor locations both work this way.
- Radius targeting. You target everyone who lives in, or is currently located in, an area a few miles across. This is what Google Ads and Meta offer natively. It is location targeting, but it is far broader than a fence drawn around a single building.
The distinction matters because most of the high-precision use cases people read about, such as targeting everyone who visited a rival dealership last month, need the second option, and that usually means a programmatic platform rather than the self-serve ad managers.
How Geofencing Works Under the Hood
When an app on someone's phone has location permission and participates in an ad exchange, the ad request it sends can include the device's latitude and longitude plus a mobile advertising ID. A demand-side platform (DSP) compares those coordinates with the fences you've drawn. If they match, the device ID is either bid on immediately or saved to an audience segment.
Two practical consequences follow from that mechanism:
- Only a slice of visitors are reachable. People have to be using an ad-supported app with location sharing switched on while inside your fence, and since Apple's App Tracking Transparency prompt arrived, many iPhone users have opted out of tracking. Expect to capture a minority of real visitors.
- Location accuracy varies. GPS signals indoors and in dense urban areas can drift by tens of meters, so a fence around a small storefront will pick up passers-by and the coffee shop next door. Polygon fences drawn around the building footprint help, but they don't eliminate the noise.
Plan around smaller, noisier audiences than the pitch deck suggests.
Where Geofencing Ads Make Sense
Geofencing works best when physical presence at a place says something strong about buying intent, and when the purchase is valuable enough to justify a premium CPM. The strongest use cases share those traits.
Conquesting competitor locations
Fence your competitors' stores, showrooms or offices and retarget the visitors with a reason to compare. Car dealers, furniture retailers, mattress stores and gyms do this heavily, because a person who walked into a competitor is clearly in-market. Keep the creative factual: a price comparison, a trade-in offer or a test-drive incentive works better than a jab at the rival.
Events and trade shows
B2B marketers can fence a convention center for the dates of an industry event, then retarget attendees for 30 to 60 days with a meeting offer or a demo. This is one of the few ways to reach a qualified B2B audience on mobile without a list. Pair it with LinkedIn ads for B2B for the same event and you can hit attendees by job title and by location.
Local service businesses
Home services, dental practices and legal firms can fence neighborhoods that match their best customers, or places that signal a need (a new housing development, a hospital for a personal injury firm, subject to the privacy rules below). Geofencing should sit alongside search, not replace it, because someone typing "emergency plumber near me" is still a stronger signal than someone walking through a neighborhood. Our guides to pest control marketing and local SEO for dentists cover the search side of that equation.
Your own locations
Fencing your own store builds an audience of recent visitors you can retarget with loyalty offers, reviews requests or cross-sells. It's essentially remarketing for the physical world, and it complements the website-based Google Ads remarketing lists you probably already run.
Where it usually fails
Low-ticket purchases with thin margins rarely justify geofencing CPMs. Neither do businesses whose customers don't visit physical places in a predictable way, such as most pure ecommerce brands. If a $20 average order has to cover a $12 CPM and a 0.3% click-through rate, the math collapses fast.
Platform Options
Your choice of platform determines what kind of fence you can actually build.
| Platform type | Fence precision | Past-visitor audiences | Typical buyer |
|---|---|---|---|
| Google Ads (radius targeting) | Broad, roughly a mile or more | No | Local advertisers already on Google |
| Meta Ads (radius targeting) | Broad, roughly a mile or more | No | Retail, events, local services |
| Programmatic DSPs with location data | Building-level polygons | Yes | Agencies, multi-location brands |
| Managed geofencing vendors | Building-level polygons | Yes | SMBs who want it done for them |
Google Ads and Meta are fine for radius campaigns around a store or service area, and they're cheap to test. They won't let you target "everyone who visited this specific building last Tuesday." If you need display reach through Google's inventory, the Google Display Network guide explains placements and bidding.
Programmatic DSPs are where true polygon fencing and look-back audiences live. Many require minimum spends, so smaller advertisers often buy through an agency. If you're evaluating partners, our guide to choosing a programmatic advertising agency covers the questions to ask about data sources and fees.
Managed vendors bundle the data, the DSP and the reporting. Ask which data supplier they use and whether you can see placement reports.
What Geofencing Ads Cost
Geofencing is almost always bought on a CPM basis. Rates vary by vendor, market, inventory quality and how much data is layered on, but most quotes fall within a few broad tiers:
- Self-serve radius targeting on Google or Meta: whatever the auction dictates for that area, often in the single-digit to low-teens CPM range for display and social placements.
- Programmatic geofencing through a DSP: commonly quoted from the high single digits to the mid-teens per thousand impressions, before agency fees.
- Managed vendor packages: frequently sold as monthly retainers with a minimum impression commitment, which can push the effective CPM well above the DSP rate.
Plug your expected budget and rate into our CPM calculator to see how many impressions you'll actually buy. For small fences, the bottleneck is often supply rather than budget: a single store might generate only a few thousand reachable devices a month, and spending more just raises frequency.
That's why frequency caps matter. Showing the same visitor 40 ads in a week doesn't make them 40 times more likely to buy. Start with a cap of three to five impressions per device per day and adjust based on results.
Building a Geofencing Campaign Step by Step
1. Define the outcome first
Decide what success looks like before you draw a fence: offer-code redemptions for retail, booked meetings for B2B, calls and form fills for a service business. Otherwise you'll end up judging the campaign on click-through rate, which says little about revenue.
2. Draw tight fences
Use polygons around building footprints rather than circles around addresses. Exclude parking lots shared with other businesses where you can. For events, limit the fence to the venue itself and the event dates; fencing the surrounding hotels adds volume but dilutes intent.
3. Set the look-back and retargeting windows
For a competitor conquesting campaign, a 30-day look-back captures people early in a decision. For events, start retargeting the day after and run for about four to eight weeks. Shorter purchase cycles need shorter windows.
4. Match creative to the context
Someone who was just at a competitor's store needs a reason to switch. Someone who attended your trade show booth needs a follow-up. Generic brand ads waste the targeting you paid for. Write two or three variants per audience and rotate them.
5. Send traffic somewhere specific
Point ads at a landing page built for the audience, with the offer from the ad repeated at the top. A homepage throws away the context. Track that page's performance with our conversion rate calculator so you can compare it fairly against your other campaigns.
Measuring Results Without Fooling Yourself
Geofencing vendors love to report "store visits" or "foot traffic conversions." These are attributed when a device that saw an ad is later detected at your location. The metric is useful, but it has two built-in problems:
- It overcounts people who would have come anyway. Regular customers who live nearby will see your ads and visit, and the report will credit the ad for visits that were going to happen regardless.
- It inherits all the location noise. A device detected in the parking lot, or at the shop next door, can count as a visit.
The fix is incrementality testing. Ask the vendor to hold out a random control group that matches the exposed audience but doesn't see ads, then compare visit rates between the two. The lift between groups is what the campaign actually caused. If a vendor won't run a holdout, treat their visit numbers as directional at best.
For online outcomes, use the same discipline you'd apply to any paid channel. Put proper conversion tracking in place, tag every link with UTM parameters, and use unique offer codes for in-store redemptions. Then calculate return on ad spend with our ROAS calculator and compare it with the profit you need. If your margins are thin, run the numbers through the break-even calculator before you scale, because a campaign can show a positive ROAS and still lose money once cost of goods is included.
Geofencing also tends to assist conversions that close through other channels. Someone sees a conquesting ad on their phone, then searches your brand name a week later and converts through paid search. A multi-touch model in your marketing attribution software will show that path; last-click reporting will hide it.
Privacy and Compliance
Location data is the most heavily scrutinized data in advertising, and the rules are tightening. A few areas need particular care:
- Sensitive locations. Fencing healthcare facilities, reproductive health clinics, places of worship, shelters and similar sites carries real legal and reputational risk. Washington State's My Health My Data Act specifically prohibits geofencing around facilities that provide in-person health services, and US regulators have taken enforcement action against data brokers over the sale of sensitive location data. Many DSPs now block these categories by default. Don't try to work around it.
- State privacy laws. A growing number of US states treat precise geolocation as sensitive personal data that requires opt-in consent. Your vendor carries most of that burden, but you should confirm in writing how their data was collected and consented.
- UK and EU rules. Under GDPR and the UK's equivalent, precise location data needs a lawful basis and clear consent. Geofencing is still possible, but the reachable audience is smaller and the paperwork is heavier.
- Creative tone. An ad that says "We saw you at our competitor's store yesterday" feels invasive. Write ads that would make sense to anyone.
Is Geofencing Worth It for Your Business?
Geofencing ads earn their place when three things are true: physical visits signal strong intent in your category, your average order or customer value is high enough to absorb a premium CPM, and you're willing to measure incrementally rather than trusting vendor visit reports. Car dealers, home improvement, furniture, healthcare providers within the rules, multi-location retail and event-driven B2B companies tend to tick all three boxes.
If you're unsure, start with a small test. Pick one competitor conquesting audience or one upcoming event, cap the budget at what you'd be comfortable losing, insist on a holdout group, and compare the cost per incremental outcome with your best existing channels. That one test will tell you more than any case study.
FAQ
How much do geofencing ads cost?
Most geofencing ads are bought on a CPM basis. Programmatic rates are commonly quoted from the high single digits to the mid-teens per thousand impressions, and managed packages often carry monthly minimums. Small fences can also be supply-limited, so a bigger budget may only raise frequency.
Can you do geofencing with Google Ads?
Google Ads offers radius targeting around an address, which works for local campaigns but is much broader than a fence around a single building. It doesn't let you build audiences of devices that visited a specific location in the past. For that, you need a programmatic platform with location data.
What is the difference between geofencing and geotargeting?
Geotargeting reaches people based on a broad area, such as a city, ZIP code or radius, usually where they live or are currently located. Geofencing draws a tight boundary around a specific place and targets devices that enter it, either in real time or by retargeting them later.
How do you measure geofencing ad performance?
Track the outcome you care about, such as store visits, calls, form fills or offer-code redemptions, and compare an exposed group with a holdout group that didn't see ads. The difference between the two is the incremental lift. Vendor-reported foot traffic without a control group usually overstates results.