Key takeaways
- Decide account structure before anything else. Centralised, franchisee-owned and hybrid each have a real failure mode, and switching later is expensive.
- Two locations of the same brand bidding on the same term is a self-inflicted cost, and it is the default outcome when franchisees run their own accounts unsupervised.
- Radius targeting around a pin is usually wrong. Territories are defined by boundaries and drive times, not circles, and the mismatch shows up as spend outside the service area.
- Send paid traffic to the location page, not the homepage. The single most common and most expensive landing page error in multi-location paid search.
- Budget floors matter more than budget size. A location spending too little to gather data will underperform indefinitely and never reveal why.
Paid search is the channel where franchise structure causes the most self-inflicted damage. The mechanics are not especially hard; the difficulty is that dozens of independent businesses are buying the same keywords, in overlapping geography, under one brand name, often without knowing what the others are doing.
Get the structure right and it is one of the fastest channels available to a new location. Get it wrong and the system spends more each year to buy the same clicks from itself.
The structural decision
Everything else follows from who owns the accounts. There are three viable models and each fails in a characteristic way.
| Model | How it works | Characteristic failure |
|---|---|---|
| Centralised | One account, campaigns per location, funded locally or from the fund | Slow to respond to local conditions; franchisees feel they have no control |
| Franchisee-owned | Each owner runs their own account or hires a local agency | Locations bid against each other; brand terms bought at a premium; no shared learning |
| Hybrid | Brand terms and structure central; local budget and some campaigns local | Boundary disputes about what is central; requires real governance to hold |
Hybrid is where most mature systems land, and it works when the split is written down rather than negotiated case by case. The usual line: the centre owns brand terms, account structure, conversion tracking and negative keyword hygiene; the location owns budget level, local promotions and service-line emphasis.
Fully franchisee-owned is defensible in systems where territories do not overlap and owners are genuinely sophisticated. It is rarely defensible in a dense metro, for the reason in the next section.
Cannibalisation is not theoretical
When two locations of the same brand bid on the same keyword in overlapping geography, the auction does what auctions do. You are not competing with a rival; you are raising your own cost per click and splitting your own impression share.
This happens by default, not by mistake. A franchisee in a neighbouring territory targets a radius that crosses your boundary because that is what the interface makes easy. Neither of them can see the other's account. Both conclude that the market is more competitive than it was last year.
What actually prevents it:
- Geo-targeting defined by the territory, not by a radius. Boundaries as a list of postcodes or a drawn polygon matching the franchise agreement
- A written rule about overlap zones where territories genuinely abut, naming which location bids and how leads from that zone are handled
- Shared negative keyword lists, including location modifiers for neighbouring territories, distributed centrally and required in every account
- Brand term ownership held centrally. Your own brand name is the term most often bid up by your own locations, and it is the one with the clearest argument for central control
- Some visibility across accounts. Even a monthly report showing where spend landed geographically catches most of this
Geo-targeting that matches reality
A radius around a pin is the default and it is almost always wrong for a franchise territory. Territories are polygons defined by an agreement; a circle either spills outside the area the location may serve or excludes part of it, and usually both at once.
Three refinements that matter more than they appear to:
- Use drive time rather than distance where the service requires the customer to travel. Five miles across a river is not five miles
- Set location targeting to presence, not interest. The default in most platforms includes people merely showing interest in the area, which for a local service business means paying for someone in another state researching a move
- Exclude aggressively at the boundary rather than relying on the inclusion shape alone, particularly where a neighbouring territory is held by another owner
The territory mapping work behind this — where boundaries genuinely fall, where media coverage does not match them, and what to do where they disagree — is covered separately in territory mapping and local media overlap.
Landing pages: send them to the location
The most expensive mistake in multi-location paid search is sending a geo-targeted click to the brand homepage. The visitor searched for a service near them, clicked an ad naming their town, and arrived at a national page that asks them to find their nearest location.
Paid traffic should land on the page for the specific location, with that location's address, hours, phone number and reviews visible without scrolling. This is the same asset that carries your local organic visibility, which is one reason location pages repay investment twice — the structure of those pages is covered in multi-location local SEO.
Where a location has a genuinely distinct offer — a promotion, a service line the others do not run — a dedicated landing page is worth building. Where it does not, the location page is better than a thin duplicate, because it carries the trust signals a purpose-built landing page usually strips out.
Budget: floors, not fairness
Franchise budget conversations tend to focus on equity — every location contributing proportionally. The more useful consideration is whether each location's budget is above the threshold where the campaign can function at all.
A campaign spending below that floor never accumulates enough conversion data for automated bidding to work, shows intermittently, and produces results too noisy to learn from. The location concludes that paid search does not work in their market. It is not possible to tell from the data whether that is true.
Where the floor sits depends entirely on the vertical and the market — a competitive metro in a high-value service category needs a great deal more than a rural territory in a low-cost one. The rule that transfers is that it exists, that it should be calculated rather than guessed, and that it is better for a system to have twenty locations funded above the floor than forty funded below it.
The new location exception
A location that has just opened has no reviews, no local organic presence and no word of mouth. Paid search is the only channel that works on day one, and the case for a temporary elevated budget — often from the fund or from opening support rather than from the owner — is strongest here. That is part of the wider opening plan in grand opening marketing.
Tracking that survives the structure
Attribution in a multi-location system breaks in specific ways, and they are worth designing around rather than discovering.
- Calls are the dominant conversion in most local service franchises, and call tracking numbers must be handled carefully so they do not contradict the number on your Google Business Profile and location page markup
- Conversion definitions must be identical across locations or no comparison between them means anything. This is the most common reason location benchmarking produces nonsense
- Lead routing from overlap zones needs a stated rule, or two locations will both claim the same conversion and the totals will not reconcile
- Offline outcomes matter more than form fills in high-value verticals. A system that optimises to leads rather than to jobs will buy the cheapest leads available, which are rarely the best ones
If locations are going to be compared on paid search performance — and they will be — the definitions have to be settled first. The politics of that comparison are covered in benchmarking locations.
The order to fix things in
If an existing multi-location paid search program is underperforming, the sequence that finds the most money fastest is fairly consistent: check where spend is landing geographically, then check what brand terms are costing and who is bidding on them, then check where paid traffic is landing, then look at budget floors. Campaign-level optimisation comes after all four.
Most systems find something structural in the first two steps, and it is usually larger than anything a bid adjustment would have recovered.
Questions we get asked
Should franchisees run their own paid search accounts?
It depends on territory density. Where territories genuinely do not overlap and owners are sophisticated, franchisee-owned accounts can work. In dense metros it reliably produces locations bidding against each other on the same terms. Most mature systems use a hybrid: central control of brand terms, structure and tracking, with local budget and promotions.
How do you stop franchise locations bidding against each other?
Define geo-targeting by territory boundaries rather than a radius around a pin, publish a rule for overlap zones naming which location bids, distribute shared negative keyword lists including neighbouring location modifiers, and hold brand term campaigns centrally. Radius targeting is the usual root cause.
Where should local paid search ads send traffic?
To the specific location's page, with its address, hours, phone number and reviews visible immediately — not to the brand homepage. Sending a geo-targeted click to a national page that asks the visitor to find their nearest location is the most common and most expensive landing page error in multi-location paid search.
Should franchise paid search use radius targeting?
Usually not. Franchise territories are polygons defined by an agreement, so a circle will both spill outside the permitted area and exclude part of it. Target by postcode list or drawn boundary, use drive time where customers travel, and set targeting to presence rather than interest.
What is a minimum viable paid search budget for a franchise location?
It varies entirely by vertical and market, but the threshold exists and should be calculated rather than guessed. Below it, a campaign cannot accumulate enough conversion data to bid effectively, shows intermittently, and produces results too noisy to learn from — and the location wrongly concludes the channel does not work.
Want this mapped to your system?
We will walk your recruitment funnel and your franchisee marketing on one call, and show you where it is leaking. No obligation.
Book a Demo