Multi-location SEO and how map pack rankings are actually decided

Everything in the previous lessons scales linearly until it does not. One Business Profile is a task. Twenty are a process, and the failure modes change completely - from "is this field filled in" to "is this consistent across twenty things maintained by different people".

The map pack is not one ranking

The most important thing to understand about map pack rankings is that there is no single ranking to hold. Results shift with the physical position of the searcher, sometimes street by street. You are not ranking third for "dentist manchester" - you are ranking third from one point, first from another, and nowhere from a third.

This has two practical consequences. First, checking your ranking from your office tells you almost nothing, because your own location biases the result. Second, the correct measurement for local is a grid across your service area rather than a single position, which is what local rank tracking tools are built to produce.

It also sets a hard ceiling. Distance is a real ranking signal, so there are searches inside your city you cannot win from where you are, no matter how good your profile is. The right response is to stop optimising for them and concentrate on the radius you can actually hold.

What actually moves a map pack position

Once distance is fixed, the levers in rough order of impact:

• Primary category - still the biggest single relevance lever, and still the most common thing set wrong.
• Review signals - count, recency, velocity, and the words in the reviews.
• Prominence from links and mentions - especially locally relevant ones.
• Profile completeness - services, hours, attributes, photos.
• The linked website - its authority, and how well the linked page matches the query.
• Behavioural signals - clicks, calls, direction requests. You cannot fake these, but a better profile earns more of them.

Note where "publish more blog posts" sits: nowhere. Content wins organic local queries, not map pack positions.

One profile per real location, and no more

The rule for multi-location is strict and worth stating plainly: a Business Profile requires a genuine, staffed location with real signage that customers can visit, or a properly configured service-area setup with no fake address.

Virtual offices, co-working desks used as address cover, and mailbox addresses are the most heavily policed form of local spam. Competitors report them, Google removes them, and the removal often takes legitimate nearby listings down with it during the review.

Where a business genuinely has multiple departments at one address - sales and service, for instance - separate profiles are sometimes permitted, but the bar is a genuinely separate operation, not a separate marketing objective.

Location pages at scale

Every location needs its own indexable page, and every one of those pages needs its own reason to exist. The scaled version of the problem from lesson three is the same problem with a bigger blast radius: twenty templated pages that differ only by town name will be treated as one page repeated twenty times.

The parts that must be unique per location:

• Title tag and meta description.
• The opening body copy - written, not merged from a template.
• Staff, photos, and the actual premises.
• Reviews and testimonials from that location.
• Hours, parking, access, and anything else genuinely local.
• Its own LocalBusiness structured data block.
• Embedded map and directions for that address.

Meta descriptions are where scaled location pages give themselves away first. A CMS that generates them from a single pattern produces twenty near-identical descriptions, which both suppresses click-through and confirms to a crawler that the pages are variants rather than distinct entities.

Structure and internal linking

Use a consistent, shallow URL pattern - /locations/manchester/ - and publish a locations index that links to every one of them. That index is what gets crawled and what feeds discovery for new locations.

Link each location page to the services offered there, and link service pages back to the locations that offer them. What you should not do is cross-link every location to every other location in a footer block: it dilutes internal signals and looks exactly like what it is.

The governance problem

At scale, local SEO stops being an optimisation problem and becomes a data governance one. Hours change locally. A branch manager edits a profile. A location moves and nobody tells marketing. Six months of that produces a set of profiles that quietly disagree with the website, the schema, and each other.

What holds it together is boring: a single source of truth for every location record, a scheduled audit of profiles against it, controlled access so edits are visible, and a defined process for openings, closures, and moves. Closures in particular are handled badly almost everywhere - a closed location should be marked permanently closed on the profile, not deleted, and its page should redirect to the nearest live location rather than 404.

That closes the Local SEO pillar. Category and NAP accuracy get you eligible, unique location assets get you relevant, and reviews and local links get you prominent - and at scale, the thing that keeps all three true is process rather than tactics.