SEO and Strategy
The Problem With Writing Healthcare Content for Search Engines
By Miracle Olajuyigbe 5 min read
Someone types this into their phone at 11:40 at night:
chest pain left side comes and goes
Now, what did they actually want?
Not a definition of angina. Not a list of fourteen possible causes ranked by prevalence. Not a paragraph explaining that chest pain is a common presenting complaint in primary care.
They wanted to know whether they can go to sleep.
That is the whole problem in one line. Search gives us the words somebody typed. It does not give us the state they were in when they typed them, and in healthcare the state is most of the information.
I write content that ranks. I have run marketing operations with revenue attached, so I am not romantic about traffic. But healthcare is the one category where optimising for the search engine and serving the human being come apart most sharply, and pretending otherwise produces pages that perform well and help nobody.
Here is where the conflict shows up.
The format fights the message
A lot of what ranks well is structurally flat. Ten symptoms, each with a heading, each about the same length, each with a tidy closing line telling you to see a doctor.
Clinically, those ten things are not equal. One of them means call an ambulance now and three of them mean mention it at your next appointment. But the format gives them identical visual weight, so the reader scans down and everything blurs into one uniform level of mild concern.
The page is accurate. The format has quietly stripped out the urgency ranking, which was the most clinically useful information available.
You can fix this, and it usually costs you a little of the neat symmetry that scanners and search engines both like. A red-flag box that breaks the pattern. A section on what to do tonight versus what to do this month. Three items instead of ten, if three is the honest answer.
Optimising for time on page is backwards here
Most content strategy treats a reader leaving quickly as failure. Engagement, dwell time, scroll depth, pages per session. For a lot of health queries, the ideal outcome is the reader leaving your page within thirty seconds because they now know they need to phone someone. That is a successful page. It will look like a bad one in your analytics.
The reverse is also true. Someone reading your symptom article for nine minutes at midnight is a wonderful engagement number and quite often a person spiralling, who should have been given a clear answer in the first paragraph and sent to bed.
If you are writing in this space, you have to decide in advance which metric you are willing to lose. Otherwise the incentives will quietly make the decision for you, one A/B test at a time.
Answering for the snippet is not answering the question
There is enormous pull toward giving a crisp forty-word answer near the top, because that is what gets pulled into the featured position at the top of results.
Sometimes that is genuinely good writing. Get to the point, respect the reader’s time.
Sometimes the honest answer does not compress. “Is this chest pain serious?” has no forty-word answer that is both accurate and safe. Any version short enough to win the snippet is either falsely reassuring or needlessly frightening, and the reader may never scroll past it to reach the nuance that made it responsible.
When a question does not have a short answer, the useful move is to say so immediately and then give the shortest honest one. “It depends on what it feels like and what comes with it. Here is how to tell the difference, and here is what means go now.” That is still a direct opening. It just refuses a compression that would have cost accuracy.
Keyword phrasing is not human phrasing
People search in fragments. “A1c normal range chart,” “metformin side effects how long.”
Writing those strings into your headings verbatim, over and over, produces prose that reads like it was assembled rather than written, and readers notice. They may not be able to name what is off, but they trust it less.
Use the phrasing to understand what people want to know. Then write in sentences a person would actually say. Search engines have been getting better at meaning rather than string-matching for years now, and the pages that read like a knowledgeable person talking tend to do fine.
“Consult your doctor” is usually an escape hatch
It appears at the end of nearly every health article, and most of the time it is doing no work at all.
The reader is often searching precisely because seeing a doctor is difficult. It costs money, or a day of work, or a six-week wait, or a two-hour journey. Telling them to consult a professional, with no further guidance, hands the problem back unchanged and lets the writer feel responsible while being useless. The version that helps is specific. Which kind of doctor. How urgently. What to say when you get there. What to bring. What would make it an emergency department instead of an appointment. That is real advice, it is safer than the vague version because it actually routes people correctly, and almost nobody writes it because it takes clinical knowledge to produce.
Where the two goals do agree
I do not want to overstate the conflict, because a lot of good SEO advice is just good writing advice.
Answer the question early. Use clear headings so people can find their section. Write for the actual intent behind the query rather than the words in it. Keep it readable. Cover the topic properly instead of thinly. Show why you are worth believing.
All of that helps rankings and helps humans at the same time. The tension only appears at the edges, and it appears in health more often than elsewhere because the stakes of a misread page are higher than in almost any other category.
The question worth asking
Before publishing, I ask what the most vulnerable plausible reader is doing thirty seconds after they land.
Not how long they stayed. Not whether the target phrase appears often enough. What they do next, and whether it is the right thing.
Sometimes that means writing a page that will never win the snippet, or one that sends people away faster than your analytics would prefer. That is an acceptable trade, and in this category I think it is the only defensible one.
Traffic is worth having. It is just not the thing you are actually being paid to produce.