How to Structure Paid Media Accounts for Multiple Locations

Structure follows the questions you need answered

An account structure is not a filing system. It decides what you can control, what the platforms can learn and what your reports can tell you. So before anyone rebuilds anything, write down the questions leadership needs the account to answer.

For most organizations with several locations, those questions are some version of:

• What does each location spend, and what does it get for it?
• Can we push one location harder without starving the others?
• Are our locations competing with each other for the same searches?
• Which services sell well everywhere, and which only in certain markets?
• Who is allowed to change what?

Each question implies a structural choice. If you need to control spend per location, budgets have to be set per location somewhere in the account. If you need to compare services across the network, services must be named and grouped the same way everywhere. Structure that cannot answer the questions will be worked around with spreadsheets, and spreadsheets are where accuracy goes to retire.

This article covers the paid media mechanics. The wider strategy (what stays central, what goes local, who governs) is set out in marketing strategy for multi-location organizations, and the account should follow from it.

One account or many

The default should be one advertising account per platform, owned by the organization, with access granted to staff and suppliers rather than accounts owned by them. Google Ads manager accounts and Meta's business-level ownership both exist to make this workable: the organization holds the keys, and people are added and removed as roles change.

Separate accounts per location make sense in a narrower set of cases:

• Different legal entities or franchisees who pay their own media and need their own invoices.
• Genuinely separate businesses under one owner, with different brands, customers and websites.
• Billing rules that require spend to land in a particular cost centre and cannot be met with budgets inside one account.

Even then, link the accounts under one manager or business portfolio so the organization keeps visibility and ownership. The failure to avoid is the common one: a location manager opened an account years ago on a personal login, connected a personal card, and is now the only person who can see the history. If that sounds familiar, the ownership checks in the paid media audit checklist for leadership teams are the place to start.

Campaign splits: control versus signal

The central trade-off in multi-location structure is between control and signal. Splitting campaigns by location gives you control: separate budgets, separate bids, separate ads. Consolidating gives automated bidding more conversions to learn from in one place. You cannot maximize both, so decide deliberately.

Split by location when

• Budgets must be controlled per location, because each location funds its own media or has its own targets.
• Demand differs sharply: one location is in a city, another in a town where a handful of people search for your service each week.
• Offers, hours, prices or services differ by location in ways the ad needs to state.
• The market speaks a different language, as with French-language campaigns for northern and southeastern New Brunswick.

Consolidate when

• Individual locations produce too few conversions for automated bidding to learn from on their own.
• Locations offer the same thing at the same terms and you are happy for spend to flow to wherever demand appears.
• The main reason for splitting is that it seemed tidy.

The usual answer is a hybrid. Campaigns are organized by intent (brand, core services, research-stage searches) as described in Google Ads management for Atlantic Canadian businesses, then split by location only for the markets large enough or different enough to justify it. Smaller locations are grouped into a shared regional campaign with location-level reporting rather than location-level budgets.

Location intent and geography

Multi-location accounts waste money in predictable ways when geography is set loosely. The settings that matter are not exotic, but they need a rule applied consistently.

• Define each location's trade area on purpose. A radius around the address is quick. Boundaries that match how customers actually travel (a municipality, a set of postal areas, a commuting zone) are usually better. Write the definition down.
• Decide how to handle overlap. Where two locations' areas overlap, either assign the overlap to one location or exclude it from both campaigns and let a shared campaign cover it. Leaving it to chance means your own campaigns enter the same auction and push each other's prices up.
• Target people in the area, not people interested in it, unless you want visitors from away. Google's broader default suits national advertisers; for a clinic or a branch office it quietly spends money on people who will never drive there.
• Handle searches that name a place. Someone in Moncton searching for your service in Saint John wants the Saint John location. Campaigns for each location can include place-name searches regardless of where the searcher is, as long as the ad and page are clearly about that location.
• Connect your Business Profiles. Linking Google Business Profiles to Google Ads lets ads show the nearest address and directions. It only works if every profile is owned by the organization and accurate, which is the subject of local SEO for multi-location organizations.
• Route the click and the call to the right place. Each location campaign should land on that location's page, and calls should reach that location's phone (or a central line that knows which location the caller wanted).

Shared signals and shared rules

Whatever you split, some things must be identical across every location, or nothing can be compared and the platforms learn inconsistent lessons.

• Conversion definitions. A "lead" means the same thing at every location: the same form, the same call length threshold, the same booking event. One location counting page views as conversions will look like the star of the network until someone checks.
• Conversion values, where you use them, set by the same method everywhere.
• Negative keyword lists managed centrally and applied to every campaign: job seekers, free versions, unrelated meanings of your service name.
• Audience lists and exclusions, such as existing customers or staff, shared rather than rebuilt per location.
• Naming conventions for campaigns, ad groups and tracking tags, so that location, service and intent can be read from the name and filtered in reports. For example: platform, location code, service, intent. Boring, essential.
• Offline outcomes from your CRM, sent back to the platforms with the location attached, so bidding learns which locations produce real customers. The mechanics are in how to measure lead quality from paid campaigns.

Shared signals also let small locations benefit from big ones. A consolidated conversion action, fed by every location, gives the bidding system enough data to work even where one branch alone would not.

Allocating budget across locations

Once the structure exists, someone has to decide how money moves between locations. Three approaches are common:

1. Equal shares. Simple and politically easy. Usually wrong, because markets differ in size and competition.
2. Proportional to revenue or capacity. Better. Locations with more chairs, beds, trucks or staff get more budget because they can serve more customers.
3. Proportional to opportunity. Best, when the data allows. Money goes where the next increment buys qualified customers at or below target, within each location's capacity.

Two rules keep this honest. First, decide in advance who can move budget between locations and how often, so the loudest location manager does not win by default. Second, compare locations against their own market, not against each other. A rural location with a modest cost per qualified lead may be performing better relative to its demand than a city branch with twice the volume.

Reporting by location

Good structure pays off in reporting. Leadership should be able to see the whole network on one page, then drill into any location without a different report or a different definition.

A location report should show, for each location and for the total:

• Spend, inquiries, qualified inquiries and (where the CRM allows) customers won.
• Cost per qualified inquiry against that location's target.
• Share of available search demand, where it can be estimated, to show whether a location is limited by budget or by its market.
• A short note on anything unusual: a staffing gap, a closure, a local competitor opening.

That requires the CRM to record which location each lead belongs to, captured automatically from the page or campaign rather than typed by hand. The wider measurement design is covered in how to measure marketing performance across multiple locations.

A worked example

An illustrative case: a physiotherapy group with five clinics, two in Halifax, one in Dartmouth, one in Truro and one in Moncton. Each clinic manager has run ads at some point. There are three Google Ads accounts, one of them on a former employee's login.

A sensible rebuild looks like this:

1. Move every account under one organization-owned manager account, and close or archive the orphan.
2. Agree one definition of a booked assessment, tracked identically on every clinic page and in the booking system.
3. Run one branded campaign for the whole group, one core services campaign for the two Halifax clinics and Dartmouth (with trade areas drawn so they do not overlap), and separate campaigns for Truro and Moncton, where demand and competition differ.
4. Share negative keywords and existing-patient exclusions across all campaigns.
5. Report every clinic on one page, each against its own target, reviewed monthly by the operations lead who decides on budget moves.

Nothing in that list is clever. The gain comes from being able to see the network clearly, so decisions about money are made on evidence rather than on which manager asked last. Building that kind of architecture across Google, Meta and LinkedIn is the first piece of our paid advertising work with multi-location clients.