Finding the questions patients actually ask

Four sources, none of them paid

Keyword research in the aesthetic sector has a specific problem: the highest-volume terms are frequently brand names you may not build pages around, for the reasons in Module 05, The rules that constrain everything you publish, and the tools that report volume have no idea about that. A clinic that starts from a volume list therefore starts from a list of pages it should not build.

The four sources below cost nothing, produce questions rather than fragments, and carry the context that tells you whether a question is worth a page.

Source one: your own clinic

Every consultation contains questions. Almost no clinic records them, and the ones that do discover within a month that the same eight questions account for most of what patients want to know, and that the website answers two of them.

The method is deliberately low effort, because anything requiring effort will stop within a fortnight. A sheet at reception and one in each consulting room. One line per question, in the patient's words rather than paraphrased. No categorisation at the point of capture.

Your enquiry inbox is the second half of this source and it is easier, because it is already written down. Read three months of enquiries and note every question asked before booking. Those are, by definition, the questions your website failed to answer.

Source two: Search Console

The performance report tells you which queries already bring people to your site, how often your pages were shown for them, and how often they were clicked. It is the only source in this list that reports what actually happened rather than what might.

Three ways to read it usefully. First, sort by impressions and look for queries where you appear often and are rarely clicked: those are pages that match the subject but not the question. Second, look for queries that are phrased as questions, which are frequently ones you have answered accidentally in a paragraph and could answer deliberately in a page. Third, look at the queries for one specific page and ask whether the page is genuinely about all of them; a page collecting many unrelated queries is usually two pages.

What it does not tell you is what people searched for and never found you at all, which is the whole of the early journey. That gap is what sources three and four fill.

Source three: the engine's own suggestions

Search engines publish, in the interface itself, a great deal about how queries are phrased: autocomplete suggestions, related searches at the foot of the results, and the expandable question panels. None of it comes with volume figures and all of it comes with phrasing, which is the more useful half.

Work treatment by treatment. Type the treatment name and read the suggestions without clicking. Type it followed by "for", "after", "does", "how long", "can I" and read again. Record the phrasings, not just the topics. Twenty minutes per treatment produces more usable material than an afternoon with a volume tool.

Source four: the six question families

This is the systematic sweep that catches what the other three miss. For each treatment, write at least two real questions in each family: definition, suitability, process, outcome, risk and money. Twelve questions per treatment, most of which will already have appeared in your other three sources, and the ones that have not are usually the interesting ones.

The outcome and risk families are where clinic websites are thinnest, and they are where the patient's actual anxiety sits. "What if I do not like it" is an outcome question, it is asked in almost every consultation, and it appears on very few clinic websites.

Prioritising the list

You will finish with more questions than you can answer, which is the correct problem to have. Prioritise on three criteria in this order.

1. Frequency in your own clinic. A question asked in most consultations is worth a page even if no tool reports demand for it, because answering it saves consultation time and because your patients are not unusual.
2. Whether anybody local answers it well, from the gap survey in Module 08, Reading the local market without naming names.
3. Whether you can answer it with something more than a paraphrase, which is the competence test from Module 07, Choosing the treatments to compete on applied at question level.

Take the top twelve. That is Part IV of your build. The remainder become answers within existing treatment pages rather than pages of their own, and Module 16, The answer page pattern explains how to decide which is which.