A specialist opening their own practice has spent a decade becoming excellent at one thing and roughly zero time on the question of how anybody finds out. The instinct is to treat it as advertising: a website, a logo, some ads. That is the wrong model, and in Switzerland it is also the one most likely to get a complaint filed against you.
Medicine is a regulated profession with rules about what you may claim, and the demand you actually want does not arrive through advertising anyway. It arrives through referral and through search — someone in pain typing a symptom, or a GP deciding who to send them to. Those two channels reward completely different work from a campaign.
What can a Swiss doctor actually say in public?
Factual information about qualifications, specialisation, location and services. Not comparative claims, not success rates presented as promises, and not anything that could be read as promising a cure. Swiss law governs medical professions directly, and the Unfair Competition Act applies on top of it: a claim you cannot substantiate is actionable by a competitor, not only by a regulator. Write everything as if a colleague will ask you to evidence it.
The practical effect is that most of the marketing playbook a general agency will hand you is unusable. Testimonials about outcomes, before-and-after comparisons, superlatives and "leading specialist" phrasing all sit somewhere between risky and prohibited. The Federal Act on Medical Professions sets the professional duties, cantonal health authorities enforce practice licensing, and if you ever mention a product rather than a service, Swissmedic rules on therapeutic-product advertising apply as well.
This is not a constraint to work around. It is a filter that removes the tactics that were never going to work and leaves the ones that do.
Why specificity beats visibility
The reflex is to want to be seen by everyone. For a clinical practice that is both impossible and undesirable. A neurosurgeon does not need traffic; they need the small number of people each month who have a specific condition and are within travelling distance.
That changes the unit of work. Instead of one page that says what you do, you need one page per condition you actually treat, each written to answer the question a patient or a referring GP types. "Trigeminal neuralgia treatment Zurich" is a query. "Excellence in neurosurgical care" is not a query, and nothing you do will make it one.
The same logic applies to the referral side. A GP deciding where to send a patient is checking whether you handle that specific presentation, what your intake process is, and how quickly you can see someone. A page that answers those three things does more for referral volume than any amount of brand work.
What does algorithmic positioning mean for a clinic?
It means deciding what to publish based on measured demand rather than on what feels important. You pull the real search volumes and phrasings for every condition you treat, rank them by volume against how well the existing results answer them, and write in that order. Then you measure which pages produce enquiries and let that reorder the queue.
It is not a growth-hacking tactic and there is nothing automated about the writing. The algorithmic part is the prioritisation: an explicit, repeatable rule for what gets published next, replacing the usual method of publishing whatever the busiest person had an opinion about.
For a practice with limited time, that ordering is most of the value. Twelve pages written in the right sequence outperform forty written in no sequence, and they cost a fraction of the effort.
What actually produces patients, and how fast
The honest answer is that the channels differ enormously in both speed and durability, and most practices pick the fast one and then wonder why it stops when they stop paying.
| Channel | Time to first effect | What it costs when you stop | Regulatory exposure |
|---|---|---|---|
| Referring GPs and colleagues | Weeks to months | Nothing; relationships persist | Low |
| Condition pages ranked in search | Three to nine months | Nothing; pages keep ranking | Low if factual |
| Google Business Profile and directories | Days to weeks | Little; listing remains | Low |
| Paid search | Days | Everything; volume stops with the budget | Medium; ad copy is scrutinised |
| General brand advertising | Slow and hard to attribute | Everything | Highest |
Timings are typical of what we have observed in Swiss practice launches and vary with competition in the canton and the specificity of the condition.
The sequencing that follows from this table is unglamorous. Get the practice listed and the intake process working. Write the condition pages in demand order. Speak to referrers directly. Use paid search only to cover the gap while the pages mature, and treat it as rent, not as an asset.
How long before a new practice sees referrals?
Direct referrals can start within weeks, because they depend on conversations rather than on indexing. Search-driven enquiries take longer: a new domain with no history typically needs several months before condition pages rank for anything competitive. Practices that plan for both timelines at once stay solvent through the gap. Practices that plan only for search run out of patience in month three.
The part nobody budgets for
Intake. A practice can rank first, get the call, and lose the patient because nobody answers the phone between 12:00 and 14:00 and the contact form goes to an address the practice manager checks weekly.
This is the least interesting problem and the most common one. Before spending anything on being found, make the path from "found you" to "booked" work: one number that is answered, one form that reaches a person, and a stated response time you actually meet. Everything upstream is wasted if this leaks. It is the same failure pattern we describe for digital transformation in the first 90 days — the constraint is almost never the technology.
Patient data makes this stricter than it would be in another sector. Anything you collect through a form is health-adjacent personal data, which means the FDPIC rules on processing and retention apply to your website as much as to your clinical systems. If you are considering any automated triage or assistant on top of that, the boundaries are narrower still — we set them out in what AI is actually allowed to do in Swiss healthcare.
Where this fits
Most of the work above is not marketing. It is digital transformation at the scale of a small practice: making the process from first search to first appointment observable and reliable, then improving the parts that measurably leak. The clinical and regulatory context that shapes it is covered across our healthcare work.
A practice of three people does not need a strategy document. It needs to know which twelve pages to write, in what order, and who answers the phone.
Related reading
- What AI is actually allowed to do in Swiss healthcare
- Digital transformation in an SME: the first 90 days
- How to choose a management consulting firm in Zurich
- Digital transformation services
Working on this right now?
Tell us where you are in two questions. A consulting partner reviews every enquiry within 2 business days.