
Which longevity or wellness business are you building, and who pays for the testing inside it? Five answers are common in this market, and Diagnostics as a Service supports every one of them. Each one has a different customer, a different revenue shape, a different place where the cost of testing falls, and a different thing that breaks first as the customer count grows.
The model is worth deciding before the panel, because the panel that suits a concierge list is the wrong panel for an employer programme. Founders often choose the panel first, so the offer is rebuilt about a year later.
None of this is legal advice, so a lawyer working in medical or competition law should review your own contracts and your website before you launch the model you pick.
Five models cover almost everything being sold today, which are the concierge list, the membership clinic, the digital-first programme, the practitioner add-on inside an existing practice, and the employer-funded channel. They differ by who pays, how often they pay, and how much human time each customer consumes.
A concierge list serves a small number of individuals who pay for continuous personal attention, so the customer is wealthy, time-poor and buying a relationship rather than a report. Revenue is a high annual fee per person, and the diagnostics cost is a small part of what they pay.
The cost of testing falls inside the fee and it is rarely the thing that decides the margin. Human time decides the margin instead, because a concierge relationship consumes hours that cannot be reduced without the customer noticing immediately.
What breaks first is the founder. This model works beautifully until the person everybody signed up for has no calendar left, at which point growth requires hiring somebody the existing customers did not choose.
From a laboratory partner this model needs breadth above everything else, since a concierge customer expects that any reasonable marker can be added on request. Over 2,500 orderable parameters across eight modalities is the kind of catalogue that supports it, together with six collection methods so an unusual sample type is never a problem.
A membership clinic sells continuous access to a local population, which means a recurring monthly or annual fee that includes one or two panels a year. The customer is a person who lives close enough to attend in person and wants a place to go.
Revenue is predictable, so the model is fundable, though it carries the fixed costs described in the startup cost article. The diagnostics cost falls inside the membership fee, which makes panel breadth a direct margin decision rather than a clinical one.
What breaks first is capacity in the fasted morning hours, because most customers want an early appointment and the room can only hold one person at a time. A clinic that sells memberships faster than it can staff its mornings creates a waiting list out of its own success.
From a laboratory partner this model needs reliable collection logistics and a trend view across repeated panels, since the second measurement is what the membership is actually for.
A digital-first programme sells a measurement and a plan to a national audience, with collection through a kit or a partner draw location rather than through premises the operator owns. The customer is comfortable buying online and expects the whole experience inside an app.
Revenue is a subscription or a repeating package, and the diagnostics cost is the largest single line in the unit economics. That makes panel design a commercial decision as much as a clinical one, which is the subject of the panel design article.
What breaks first is support volume. A kit that a customer collects incorrectly generates several messages, a redraw and a refund conversation, so every point of friction in collection multiplies by the number of customers.
From a laboratory partner this model needs an API, webhook events and a sandbox, because the programme is software and the testing has to work inside it. A white-label customer dashboard matters here more than anywhere else, since the customer should see one brand throughout.
A practitioner add-on puts testing inside a practice that already has customers, which might be a physiotherapy practice, a nutrition practice, a pharmacy or a gym. The customer is somebody who already comes through the door for another reason, so acquisition cost is close to nothing.
Revenue is per panel rather than recurring, at least at first, and the diagnostics cost is passed to the customer directly. The margin per panel is thinner than in a membership model, though the volume arrives without a marketing budget.
What breaks first is scope of practice. Staff who are not licensed to interpret results will be asked to interpret them anyway, which is why the report wording matters so much and why the abnormal result article exists.
From a laboratory partner this model needs plain-language explanations for every marker and a clean escalation route to a licensed clinician. The gym and studio version of this model is described in its own article.
An employer-funded channel sells a workforce programme to a company, so the customer who signs is a human resources or occupational health decision maker rather than the person giving the sample. One contract reaches hundreds of employees, which is why the sales cycle is long and the volume is large.
Revenue is a contract with an annual value and a defined participation rate, and the diagnostics cost is paid by the employer per participating employee. Under Section 3 number 34 of the Einkommensteuergesetz, employer contributions to workplace health promotion are free of income tax and social security up to 600 euro per employee per year2. That exemption applies only where the contributions are granted in addition to salary already owed and meet the quality and certification requirements of Sections 20 and 20b SGB V1. A diagnostic panel is not obviously certifiable under that route, because the route is aimed at behavioural prevention courses, so treat the exemption as unavailable unless a tax adviser confirms otherwise.
What breaks first is data separation. The employer must never receive an individual result, because health data is special category data under European law and the employment relationship makes consent harder to rely on3. A programme that reports individual values back to an employer is a legal problem rather than a product feature.
From a laboratory partner this model needs strict role separation in the software, aggregate reporting that cannot identify anybody, and a data agreement the company's own legal team will accept.
Every model needs the same five things underneath it, which are accredited analysis, a collection method, a courier arrangement, reporting software and a data agreement. What differs is which of the five is the hardest requirement, and that difference should decide who you sign with.
Aniva covers all five with one contract, 25+ partner draw locations across Germany and a white-label console for the partner. The consumer membership and the draw network are German, while the partner offering itself is sold across Europe.
Compliance clauses cover GDPR with an AVV and Germany-hosted infrastructure. Analysis is performed by an accredited laboratory partner, which holds ISO 15189 accreditation and operates under RiliBAEK. The partner contract carries anti-kickback clauses referencing Section 299a StGB and Section 31 of the model professional code. Contract wording does not by itself make an arrangement lawful, because the pricing and the substance decide that. Section 299b binds the paying side, so ask any supplier how its contract addresses that half.
The business model decides the panel and the retest cadence, rather than the panel deciding the business model. A concierge list can afford breadth per person, while an employer programme needs a short panel that a thousand people can take on the same morning.
Cadence follows the same logic, and it also follows the biology of each marker, which is the argument in the retest business model article. Decide the model first, then the panel, then the cadence, because doing it in any other order produces an offer that nobody can price.
Book a 30-minute demo if you want to test your chosen model against the laboratory layer before you build it.
Last updated: 31 August 2026

Aniva handles the lab, the logistics and the report. You keep the patient relationship and your own branding.
Twenty minutes, and you will know whether it fits your setup.