AI Marketing, SEO & GEO · Location-Page Strategy
Local SEO for Swiss SMEs: How to Rank in Zurich, Zug, Basel or Geneva Without Building 100 Empty City Pages
Two businesses, the same advice, two different right answers
Illustrative scenario: A fiduciary and tax advisory partnership with staffed offices in Zurich and Basel, and a pest-control company run from a single depot near Winterthur covering Zurich, Zug, Baden and Frauenfeld, were both told the same thing by the web consultants they hired last year: build a dedicated page for every city you serve. For the fiduciary partnership, that advice was correct — each office has its own team, its own opening hours, its own client base and its own reviews. For the pest-control company, it was close to the most damaging thing they could have done to their own domain.
The difference between the two is not ambition, budget or how many cities appear on a delivery van or an invoice history. It is whether each city can supply real, checkable evidence that a search engine — and a real visitor — can verify independently of the others. The fiduciary partnership's Basel office has a lease, a receptionist and a set of Basel-specific Google reviews. The pest-control company's "Zug" is a dispatch entry: a technician who drives there when booked, using the same price list, the same team and the same guarantee as every other town on the list.
This is the question this article answers, and it is deliberately narrower than "how do I do local SEO in Switzerland." It assumes the fundamentals are already in place — a claimed and maintained Google Business Profile, consistent name-address-phone data, genuine reviews (the subject of the companion article on local SEO in Switzerland, linked at the end of this one) — and goes straight to the decision that trips up otherwise well-run Swiss SMEs: should this specific location get its own page, and if some locations qualify, how many pages does that honestly add up to. The decision is not a content-marketing preference. It is a four-part evidence test, and Google names the exact failure mode the test exists to prevent in its own published spam policy: the doorway page.
The evidence test: what a location must prove before it earns a page
A candidate location — a city, a canton, a service area — earns a dedicated page only by clearing four independent tests. None of them is about desire, target market size or how good a location "sounds" in a sales conversation.
- Staffed presence or a genuinely sustained local service base: a lease, a listed employee, a phone that is answered locally, or — for a service-area model — a real, ongoing pattern of jobs completed in that area over a meaningful period, not an aspiration to expand there.
- Independently distinguishable content: something true of this location that is not equally true of the others — a named local team, location-specific reviews that mention the branch or city by name, real local case examples, actual hours, parking or access notes that differ from the next town over.
- Genuinely separable search demand: a query pattern for this location that a shared "our services plus coverage area" page would not already satisfy — not simply the same sentence with the city name swapped into the heading.
- A named owner who will keep it current: someone accountable for updating this specific page's local facts — staff, hours, reviews — as they change. An unowned page decays into precisely the stale, near-identical content the doorway-pages policy targets.
What Google actually says about this, not what agencies assume it says
Google's Search Central spam policies define doorway abuse directly, as pages "created to rank for specific, similar search queries" that "lead users to intermediate pages that are not as useful as the final destination." One of the four listed examples reads almost as if it were written about this exact situation: "having multiple domain names or pages targeted at specific regions or cities that funnel users to one page" (Google Search Central, accessed 29 July 2026). The other three examples describe the same pattern from different angles: multiple sites with slight URL and homepage variations built to capture more of one query; pages generated purely to funnel visitors into the actually useful part of a site; and substantially similar pages placed closer to the search results than a clear, browsable page hierarchy would allow.
None of this bans location pages. It bans substituting a swapped city name for real distinguishing substance — which is exactly what a templated "service plus city" page produces at scale when nobody has actually applied the evidence test above to each one individually.
One genuine exception is worth stating plainly: a landing page built for a single paid advertising campaign and marked noindex is not a doorway-page risk, regardless of how many near-identical versions exist for different cities. The policy governs pages competing for organic search placement, not private, non-indexed pages a search engine was never meant to rank in the first place.
The alternative nobody mentions: a service-area business profile
Most conversations about Swiss local SEO jump straight from "no location pages" to "build location pages," skipping the option Google itself built for exactly the pest-control company's situation: the service-area business (SAB) designation on a Google Business Profile, for a business that "visits or delivers to customers directly but doesn't serve customers at their business address" — cleaning, pest control, trades and mobile services are Google's own examples (Google Business Profile Help, accessed 29 July 2026).
Three specifics matter in practice. First, a business that never serves walk-in customers should remove its street address entirely rather than display one nobody is meant to visit — Google's own instruction is direct: "if you don't serve customers at your business address, remove your address from your Business Profile." Second, service areas are declared as up to twenty named cities, postal codes or comparable areas, not a fixed radius drawn on a map — Google explicitly disallows setting coverage as a plain distance figure. Third, the platform's own guidance is that the outer boundary of a declared area should stay within roughly two hours' driving time of the business's actual base — for a depot near Winterthur, that comfortably reaches Zurich, Zug and Baden, but not Geneva or Lugano, however much a sales pitch might want it to.
One correctly configured profile, paired with a single on-site coverage page — a map, an honest list of served towns, and the strongest set of reviews the business actually has — does everything the instinct behind "one page per city" was reaching for, without producing a single near-duplicate indexed page.
How many location pages should you actually have?
The honest number depends on business model, not ambition, and falls into three broad bands.
- A single office or depot serving a wider area with no other staffed location: typically zero dedicated location pages beyond the existing service pages. The service-area profile plus one coverage page carries the entire local-search job — this is the pest-control company's honest answer.
- A small number of genuinely staffed branches, commonly two to four: one page per staffed branch, not one per city where clients happen to reside. A Basel client of the Zurich office does not entitle Basel to its own page simply because an invoice was once sent there — this is the fiduciary partnership's honest answer, with exactly two pages.
- A larger, genuinely multi-branch network — a franchise, a clinic group, ten or more staffed locations run to a consistent operating standard: systematic location pages become legitimate and expected, but only on a template built to pull real, per-branch data — a direct phone line, branch-specific hours, named staff, branch-filtered reviews — never a shared paragraph with only the city name swapped.
What actually goes wrong when this is ignored
Google's own spam-policy documentation treats doorway pages as a policy violation subject to both algorithmic demotion and, in clearer or larger-scale cases, manual action — a materially harder outcome than simply "these pages will not rank," up to removal of the affected pages from search results entirely.
- Keyword cannibalization, well short of any formal penalty: several near-identical pages compete against each other for the same handful of queries, and search engines frequently pick none of them clearly, because no single page carries a strong enough independent signal.
- Diluted trust signals: reviews, credentials and proof that would look convincing concentrated on one or two pages look sparse and unconvincing spread thinly across ten.
- Ongoing maintenance debt: every fabricated location page needs updates — prices, hours, staff — that realistically will not happen consistently across ten or twenty pages, and stale content is itself a weaker signal over time.
- The mirror-image mistake for genuine storefront businesses: hiding a real, walk-in address to "look more local" misrepresents business information under Google Business Profile's own guidelines and creates a mismatch between what a customer sees on arrival and what the profile states.
When you can skip this decision entirely
A genuinely single-location retail business, clinic or workshop with no expansion or franchise plans and adequate current local visibility does not need any of this. The correct number of location pages is zero, not because any test failed, but because there is only one location to test in the first place — the homepage and a well-kept Google Business Profile already do the entire job. Building a technically correct location-page architecture for a problem that does not exist is still effort spent on the wrong thing.
The decision, not the theory
Do not start by asking how many location pages a Swiss business "should" have — start by counting how many real locations actually clear the four-part evidence test. Everything that does not clear it becomes a named service area on a Google Business Profile and a line in one honest coverage page, not a page of its own. The count should be revisited only when a location's underlying facts genuinely change — a new branch opens, a depot moves — never simply because a template or a fresh SEO trend makes building more pages easy. The four-part test in this article turns that count into a specific decision for your own locations, not a generic rule borrowed from a business that looks nothing like yours.
Frequently asked questions
Does a service-area business need a street address at all?
No, if it never serves walk-in customers there. Google's own guidance is to remove the address from the profile in that case rather than display one nobody is meant to visit. Keeping a hidden or unused address adds no local-search benefit and only invites a mismatch between the profile and reality.
Will removing my address hurt my local rankings?
There is no reliable evidence that it does, and Google's guidance exists specifically for this configuration. A business that never serves walk-ins loses nothing by matching its profile to reality, because visibility for a service-area business is built on its declared service areas, not a visible storefront address.
What should I do if I already have ten near-identical city pages?
Consolidate the genuinely useful content into one strong page — or the coverage page — and use 301 redirects from the retired URLs rather than deleting them outright, so the surviving page inherits whatever authority the old ones had accumulated. Then request re-crawling of the surviving page through Google Search Console instead of waiting for it to happen on its own schedule.
Two of our offices are twenty minutes apart. Do both need their own page?
Run the evidence test on each independently rather than assuming proximity settles it. Staffed presence alone is not enough — if both genuinely have separate teams, hours and reviews, both qualify; if one is essentially unstaffed overflow space, fold it into the other office's page as a note rather than building it a page of its own.
How many city names can one page realistically target?
Fewer than most templates assume. A single, well-built coverage page can honestly list several towns alongside a map, but trying to rank the same page's title and heading for more than one or two distinct city-plus-service phrases reproduces the exact "targeted at specific regions or cities" pattern the doorway-pages policy names directly.
Is a paid landing page built for one city the same doorway-page risk?
No, provided it is marked noindex and never intended to rank organically. The doorway-pages policy concerns pages competing for organic search placement; a private campaign page that a search engine is not meant to index sits outside that concern entirely.
Does this apply differently to a canton-wide business than a nationwide one?
The mechanism is identical; only the number of genuinely staffed locations changes what counts as a flagship location. A fiduciary firm with two Swiss offices and a retail chain with forty branches are both simply running the same four-part test at a different scale.
What is the fastest way to check whether we already have a doorway-page problem?
Open the Pages report in Google Search Console and look for several city pages showing near-identical, low impressions and clicks, or simply place two of the pages side by side. If swapping the city name is the only real difference a reader can find, the pattern is already present, regardless of how the pages were originally intended.
Practical AI for your business
From idea to implementation – we show you what is concretely possible in your case.
Request a demo