Medical ad copy for doctors: compliant and specific

15 min read

Medical ad copy works when it replaces adjectives with verifiable facts: specialty, location, availability, languages, insurance and how to book. Superlatives, guarantees, fear and drug names create policy and regulatory problems. Build headlines from an approved claim library, have a clinical reviewer check the whole pool and landing page, and judge copy on honoured appointments.

Good medical ad copy does three jobs in very little space. It passes Google’s review, it stays within what Indian rules allow a hospital or doctor to say, and it gives a worried patient a concrete reason to click this ad rather than the three around it. Most healthcare ads I audit manage one of the three. They are either compliant and empty (“Quality care you can trust”) or specific and risky (“Painless knee replacement, guaranteed results”).

The way through is to replace adjectives with facts. This article covers the constraints, what to say instead, labelled example ads for a fictional Example Hospital, the assets that carry detail, and a review workflow that satisfies both marketing and the clinical reviewer. It is part of the complete guide to Google Ads for doctors in India.

Three filters every line of medical ad copy passes

I ask teams to read every headline and description through three filters, in this order:

Google policy. No prescription drug names, no unreliable or improbable claims, no exploiting illness to force urgency, no phone number in the ad text, no gimmicky capitals or punctuation. The platform side is covered in Google Ads healthcare policy: what gets disapproved and why.
Indian rules. The NMC’s professional conduct framework for doctors, the ASCI Code and the Drugs and Magic Remedies Act. What those allow a hospital to say is covered in what hospital ads can and cannot say under NMC and ASCI rules. This article does not repeat it, and none of it is legal advice.
The patient. Does this line answer something the searcher needs to decide: who, where, when, how much, how to book?

A line that passes the first two but fails the third is wasted space. A line that passes the third but fails either of the first two will not run for long, or should not.

The responsive search ad, in numbers

Every search ad is now a responsive search ad. Per Google’s responsive search ads help page, you can supply up to 15 headlines of up to 30 characters each and up to four descriptions of up to 90 characters each, plus two display path fields of up to 15 characters. Google then assembles combinations for each search.

That assembly is the part clinical reviewers need to understand. They are not approving one ad; they are approving a pool of lines that will appear in combinations nobody has seen side by side. Every headline must be safe next to every other headline and description.

Two practical points follow. Pinning a line to a position is useful for a mandatory element, such as the hospital name in headline one, but Google advises against heavy pinning. And ad strength, the rating shown while you write, is a feedback tool: Google’s ad strength help page says it is not used to calculate Ad Rank or Quality Score. Do not add risky lines just to move the meter.

Specific beats superlative

What a patient wants from an ad is information, and most of it fits in 30 characters. These are the elements I build headline pools from:

The service and specialty, in the words patients use: “Knee Specialist”, “Child Specialist”, “Full Body Check-up”.
The location, down to the neighbourhood patients recognise.
Availability: OPD days, evening or Sunday clinics, same-week appointments, but only if they are reliably true.
Languages spoken, which matters more in Indian cities than most marketers assume.
Insurance and cashless, where the unit is empanelled.
Booking channel: call, WhatsApp, online slot.
Factual credentials: the doctor’s specialty and qualifications, stated plainly, with no ranking or comparison.
Price guidance for packages, where the landing page shows it.

Each of these is verifiable. That makes it easier to approve, and it is also what persuades. “OPD till 9 pm” beats “Trusted by thousands” for a parent searching at 8 pm.

Write for the moment the patient is in

The same specialty is searched in very different states of mind, and copy should follow the moment rather than the department. A parent searching for a child specialist at night wants to know who is open now and how far away they are. A caregiver searching for cancer care is usually researching on someone else’s behalf and wants to know how to share reports and get a view quickly. Someone searching a procedure cost is comparing, and wants to see that prices are available without a sales call.

So before writing, I ask the team to name the moment for each ad group in one sentence. “Parent, evening, needs a doctor today.” “Caregiver, reports in hand, wants a second opinion.” That sentence decides which facts lead the headline pool, and it stops every ad in the account sounding the same.

Words and claims I remove on sight

These are the phrases I strike from drafts, with the reason, so teams learn the pattern rather than a list:

“Best”, “No. 1”, “top”, “leading”: comparative claims that need substantiation most hospitals cannot provide, and that ASCI scrutinises.
“Guaranteed”, “100%”, “permanent”, “cure”: improbable-result language under Google’s misrepresentation policy, and a problem under Indian rules too.
“Painless”, “risk-free”, “no side effects”: clinical claims a reviewer will rarely sign off, for good reason.
Success rates and patient numbers: they need data, methodology and context that an ad cannot carry.
Fear and urgency: “Don’t ignore chest pain”, “Act before it’s too late”. Google treats exploiting illness this way as clickbait.
“FREE”, “!!!”, odd symbols: editorial issues under Google’s editorial policy, and cheapening for a medical brand.
Drug and injectable brand names: restricted drug terms in India.

Rewrites: from risky to specific

The quickest way to train a team is to show the same idea written badly and well. These pairs are illustrative only:

“Best Knee Surgeon in Pune” becomes “Knee Specialist in Kothrud”. Same intent, no unsupported comparison, and a neighbourhood the searcher recognises.
“Painless Cataract Surgery” becomes “Cataract Consultation Today”. The clinical claim goes; the availability stays.
“Guaranteed Weight Loss Results” becomes “Meet a Bariatric Surgeon”. An outcome promise becomes a factual next step.
“Don’t Ignore Chest Pain!” becomes “Cardiology OPD, Mon to Sat”. Fear goes, the useful fact arrives.
“FREE Consultation Hurry” becomes “Book Online or on WhatsApp”. The gimmick goes; the booking channel is what actually helps.

In every case the rewrite is also shorter to approve, because it states something the hospital can verify on any given day.

Healthcare ad examples, labelled

These healthcare ad examples are illustrative only. Example Hospital is fictional, as are the details, and each would need clinical and legal review before real use. They show the approach, not copy to lift. Each headline is within 30 characters and each description within 90.

Example 1 (illustrative only): knee consultation, Example Hospital, Pune
Headlines: Knee Specialist in Kothrud | Example Hospital Orthopaedics | OPD Mon to Sat, 9 am to 7 pm | Cashless Insurance Accepted | Book by Call or WhatsApp | Knee Replacement Consultation
Descriptions: Consult an orthopaedic surgeon about knee pain and treatment options. Book a slot today. | Hindi, Marathi and English spoken. See doctor profiles, OPD timings and package details.
Display path fields: orthopaedics / knee
Use the cashless line only if it is true for this unit, and keep the OPD timing current.

Example 2 (illustrative only): evening paediatric OPD, Example Hospital, Chennai
Headlines: Child Specialist Open Late | Paediatric OPD Till 9 pm | Example Hospital Paediatrics | Walk In or Book on WhatsApp
Descriptions: Evening and Sunday paediatric OPD in Anna Nagar. Check timings and book a visit online. | Vaccination, fever and routine check-ups. See our paediatricians and their OPD days.
Display path fields: paediatrics / opd
Timings belong in the ad only if someone updates them when they change.

Example 3 (illustrative only): health check packages, Example Hospital, Lucknow
Headlines: Full Body Check-up Packages | Example Hospital, Gomti Nagar | See Tests Included and Price | Home Sample Collection
Descriptions: Compare health check packages by tests included. Prices listed on the page, no surprises. | Book online or on WhatsApp. Reports shared digitally, with a doctor consult to review.
Display path fields: health-checks / packages
The price promise holds only if the landing page really lists prices.

Notice what is absent: no superlatives, no outcomes, no fear. Notice too that every line depends on an operational fact someone must keep true. The fastest way to lose patient trust is an ad promising a Sunday OPD that closed last month.

Assets carry the detail the headline cannot

Assets, formerly called extensions, add lines and links beneath the ad and often matter more than the headlines themselves. The ones I set up for every healthcare account:

Sitelinks to the doctor profiles, OPD timings, insurance and cashless page, and location page.
Callouts for factual features: languages spoken, parking, wheelchair access, pharmacy on site.
Structured snippets listing services or specialties.
Call asset with a tracked number, scheduled for the hours the line is answered. This is where the phone number goes, never in the ad text.
Location asset linked to the Google Business Profile, so the address and distance show.

If most of your patients prefer to message rather than call, a sitelink or landing page button that opens WhatsApp often outperforms a form, a pattern I explore in WhatsApp vs web forms for hospital enquiries. Make sure someone answers it within the hours the ad runs.

Doctor profile sitelinks deserve special care, because they send people to a page that has to do the persuading. What that page needs is covered in doctor profile pages that convert, and the destination for the ad itself in landing pages for doctor consultation ads.

A review workflow that marketing and clinicians both accept

Copy review fails when it is an email thread with a screenshot of one ad. The process I use:

Keep an approved claim library. A shared sheet of factual statements, such as doctor qualifications, timings, languages and empanelments, each with an owner and a last-checked date.
Draft from the library. Marketing writes headlines and descriptions using only approved facts or clearly marked new claims.
Review the whole pool. The clinical reviewer sees every headline and description, plus the landing page, not a single rendered preview.
Check the automated features. AI Max for Search campaigns can customise ad text automatically from your landing pages and assets. If it is switched on, the pages it draws from need the same review, or the feature should stay off for sensitive services.
Record approval. Note who approved what and when, and label reviewed ads in the account.
Re-review on change. Any new line, new doctor or changed timing goes back through the same steps.

The same discipline applies to any AI-drafted copy, which I discuss in AI-generated doctor content: where the review line sits. The printable pre-launch version sits inside the Google Ads launch checklist for doctors.

Testing copy without chasing clicks

Responsive search ads test combinations automatically, and Google reports how individual assets perform. That is useful, with one caution: the metric that matters is honoured appointments, not click-through rate. Copy that promises more than the page delivers can raise clicks and lower bookings.

I test one idea at a time: availability-led lines against location-led lines, or language lines against insurance lines. Give each test enough weeks to gather appointment data, not just clicks, and keep the losing lines in a log so nobody reintroduces them. For keyword themes where copy seems weak, check first whether the problem is the keyword itself, using the intent tiers in keywords that book appointments vs keywords that waste budget.

Questions people ask

What makes medical ad copy different from other ad copy?

Medical ad copy has to pass stricter platform rules, Indian professional and advertising rules, and a clinical accuracy check, while still helping an anxious patient decide. That rules out the superlatives, guarantees and urgency tactics common in other categories. What works instead is factual detail: specialty, location, availability, languages, insurance, booking channel and plainly stated credentials, each of which can be verified.

As a doctor, can my ad mention my qualifications and experience?

Stating your specialty and qualifications factually is generally the safe core of a doctor’s ad. Rankings, comparisons with other doctors, claims about outcomes or case numbers, and self-promotional language are where problems start under professional conduct rules and Google’s misrepresentation policy. Keep it plain and verifiable, let the landing page carry more detail, and take advice on anything beyond facts. This is not legal advice.

Why can we not say best hospital or number one?

Because those are comparative claims that need substantiation most hospitals cannot provide in the form ASCI expects, and Google may treat unsupported claims as misleading. Even when a hospital has an award or ranking, ASCI has specific guidelines on how awards and rankings may be used in healthcare advertising. Factual differentiators such as evening OPD, languages or cashless acceptance are safer and usually more persuasive.

How many headlines and descriptions should we write?

Google allows up to 15 headlines of 30 characters and up to four descriptions of 90 characters per responsive search ad. I aim for a full or nearly full pool built from approved facts, because variety gives the system more to test. Quantity is not the goal, though. Every line must be safe in any combination, so a smaller pool of solid lines beats a large pool with risky filler.

As a clinical reviewer, what exactly should I approve?

The full pool of headlines and descriptions, the assets such as sitelinks and callouts, and the landing page each ad points to. Check that every statement is accurate, that nothing implies an outcome or guarantee, and that the specialty and qualifications named are correct. Remember that Google combines lines automatically, so read each one as if it could appear next to any other.

Can we put our phone number in the headline?

No. Google’s editorial policy does not allow phone numbers in ad text. Use a call asset instead, ideally with a tracked number so calls can be attributed to campaigns, and schedule it for the hours the line is actually answered. A location asset linked to your Google Business Profile adds the address. WhatsApp links can go in sitelinks or on the landing page.

Should ads mention prices?

Where the landing page shows a price or package clearly, a headline such as prices listed or see tests included can lift intent and reduce wasted calls. Avoid stating specific prices in the ad unless someone owns keeping them current, because a mismatch with the page damages trust. Google’s misrepresentation policy also expects the full cost model to be clear, so pricing needs care on the page too.

How long does copy approval usually take?

With an approved claim library and a named clinical reviewer, a new ad group can be reviewed in a day or two. Without them, approval often drags for weeks through email threads. Google’s own review usually completes within a business day. Build the library first: it makes every later review faster, because most lines reuse facts that are already verified.

As a marketing head, how do we stop copy drifting across agencies?

Give every agency the same approved claim library, the same list of banned words with reasons, and the same review workflow. Require labels in the account marking reviewed ads and dates. Audit a sample of live ads monthly against the library. When doctors join or leave, or timings change, update the library first so every agency works from the same facts.

Is AI-generated ad text safe to use?

It can speed up drafting, but it needs the same review as anything a person writes, and sometimes more, because AI tools tend to reach for superlatives and outcome language. Google’s AI Max for Search campaigns can also customise ad text automatically from your pages and assets. If you enable it, review the source pages carefully, or keep it off for sensitive services.

Do ads need to be in regional languages?

Where patients search in a regional language and the landing page and front desk can serve them in that language, yes, native-language ads usually outperform translated English. Write for how people speak, not a literal translation, and put the copy through the same clinical review. Where only some staff speak the language, say which languages are spoken rather than switching the ad language.

How do we measure whether new copy is working?

Compare versions on honoured appointments per click or per spend, not only on click-through rate. Copy that overpromises can raise clicks while lowering bookings. Test one idea at a time, allow enough weeks for appointment data to accumulate, and log the results. Google’s asset performance reporting helps, but the final judgement should come from matched enquiry and appointment data.

How does this scale across a multi-unit group?

Keep one central claim library and banned-word list for the group, with unit-level sections for local facts such as timings, doctors, empanelments and languages. Build shared ad templates by specialty that units fill with local detail. Centralise clinical review standards even if reviewers sit in units. That keeps the brand voice consistent while letting each ad speak to its own catchment.

Read my takes first in Google Search