A personalised protocol is a clinical plan built from an individual patient's intake, history, goals, and — where relevant — biomarker data, then adjusted cycle over cycle based on how they actually responded. The word "personalised" is used loosely in health marketing to mean anything from a quiz-driven product recommendation to a genuinely individualised clinical plan. The distinction matters, because the two produce different outcomes. Real personalisation is a system, not a feature.
Why "personalised" has become a diluted term
Every consumer health brand claims personalisation. The word appears on supplement labels, in weight-loss ads, in cosmetic quizzes, in wellness apps. It's used to describe experiences that range from a genuine clinician-built protocol to a five-question quiz that assigns you to one of three pre-designed product bundles.
The dilution is a marketing problem, but it's also a clinical one. When a term stops distinguishing anything, patients lose the vocabulary to describe what they actually want — which is care built around who they are, adjusted over time as they change. That's a specific thing. It has a definition. And it looks very little like most of what's sold as "personalised."
The five components of a real personalised protocol
A protocol qualifies as genuinely personalised when it meets five criteria. Fewer than five, and the term is doing more work than the underlying process.
1. Comprehensive intake that captures more than symptoms
A real personalised protocol starts with an intake designed to capture the full clinical picture: medical history, current medications, past treatments and how the patient responded to them, lifestyle factors that affect outcomes, and — importantly — what the patient is actually trying to achieve.
A five-question quiz cannot do this. A twenty-minute structured intake, reviewed by a clinician, can. The intake is not a marketing exercise. It's the foundation the entire protocol is built on.
2. Clinician review of the full picture, not algorithmic pattern-matching
Automated systems can flag interactions, screen for contraindications, and generate a first-pass recommendation. What they cannot do is weigh clinical judgement — the moment when two plausible protocols are both reasonable, and the right choice depends on subtleties in the patient's history that an algorithm won't weigh correctly.
Real personalisation requires a licensed clinician who reads the intake, understands the patient's goals, and makes the call. The algorithm can support that decision. It cannot replace it.
3. A protocol that reflects the patient's specific situation — including saying no
The most under-appreciated component of personalisation is the ability to return a decision the patient didn't want.
A provider whose protocol engine can only recommend products from its own catalogue is not personalising. It's selling. A provider whose clinician can respond with "this isn't the right protocol for you right now — here's why, and here's what we'd recommend instead" is personalising. That includes recommending less than the patient asked for, or recommending they see a specialist in person first.
Personalisation that only ever produces a "yes" is not personalisation. It's a funnel with a clinical rubber stamp.
4. Structured follow-up that captures how the patient actually responded
The first protocol is a hypothesis. What makes it personalised is what happens next.
Structured follow-up means the provider is systematically capturing how the patient responded — what worked, what didn't, what side effects appeared, what the patient's subjective experience was, and (where relevant) what changed in objective markers. That data is what turns the second cycle into something better than the first.
Without structured follow-up, every cycle is a fresh guess. With it, the protocol iterates toward what actually works for this specific patient.
5. Cycle-over-cycle adjustment based on that follow-up data
The final component is what closes the loop. The clinician reviews the follow-up data at the end of each cycle and adjusts the next protocol accordingly. Dose changes, compound substitutions, sequencing changes, or discontinuations happen because the previous cycle produced information that warranted them — not because the patient complained.
This is the component most providers skip. It's the most operationally expensive part of personalisation. It's also the one that produces the difference between year one and year three.
What personalisation is not
Some things sold as personalisation, that aren't:
A quiz that routes you to one of a small number of pre-designed bundles. That's segmentation, not personalisation.
A subscription that ships you the same product every month with your name on the label. That's fulfilment, not personalisation.
A recommendation engine that never returns "no." That's a sales funnel with a clinical layer, not a clinical process.
A "personalised" plan that never changes after the initial protocol is set. That's a static prescription with better branding.
None of these are inherently bad products. They may be exactly what a patient wants at a given price point. But they should not be described as personalised, because they're not doing what personalisation clinically means.
Why cycle-over-cycle adjustment matters more than the initial recommendation
If you had to pick one thing that separates real personalisation from marketed personalisation, it wouldn't be the intake. It would be what happens after cycle one.
The reason is clinical. No clinician, no matter how experienced, can perfectly predict how an individual patient will respond to a specific protocol on the first attempt. Human physiology has too much individual variation. The value of a personalised protocol comes from the second, third, and fourth iterations — where the clinician has actual response data from this specific patient and can refine accordingly.
A provider that spends heavily on the initial intake but has no structured process for cycle two is optimising the wrong thing. The intake is the entry. The iteration is the value.
This is also the answer to why "personalised" is often more expensive than commodity alternatives. The cost isn't in the first prescription. It's in the clinician time required to review response data and adjust every cycle. Providers who charge accordingly are pricing to the actual work. Providers who don't are usually skipping the work.
How to tell if a provider is personalising or just marketing personalisation
Practical diagnostics you can run before signing up with any provider claiming to personalise:
Ask how the initial protocol is decided. If the answer is "our algorithm," ask what the clinician does. If the answer is "the clinician reviews the algorithm's output," ask how long that review takes.
Ask what happens at the end of the first cycle. If the answer is "you re-order," that's not personalisation. If the answer describes a structured review with your clinician, that's closer.
Ask what percentage of patients receive an adjusted protocol in cycle two versus the same protocol repeated. A provider that adjusts most patients between cycle one and cycle two is doing the work. A provider that repeats the same protocol for everyone is not.
Ask what an ineligibility decision looks like — see the previous piece in this journal for why that question is diagnostic across telehealth generally.
The answers won't always be perfect. What matters is that the provider can answer at all. Vagueness is the signal.
What good personalisation costs — in money and in patience
Real personalised protocols cost more than commodity alternatives, for a straightforward reason: the clinical work behind them is more expensive to deliver. Longer intake review, structured follow-up, cycle-over-cycle adjustment, and clinician time all add operational cost that has to be priced in.
They also cost more in patience. The first cycle of a genuinely personalised protocol will not necessarily produce the outcome the patient hoped for. The protocol is a hypothesis. The value shows up in cycles two, three, and four, once the clinician has actual response data to work with. Patients who abandon after one cycle rarely see the difference personalisation makes.
If you're paying for personalisation and expecting commodity-cycle speed, you're going to be disappointed. If you're paying for commodity and expecting personalised iteration, you're going to be disappointed differently. Match your expectations to what you're actually buying.
Medical disclaimer
This article is general education, not medical advice, and is not a recommendation for any specific medication. Consult with a licensed healthcare provider to determine what treatment is appropriate for you.
Clinically reviewed by
Amelia Baweja
