Healthcare PPC keywords that book appointments

15 min read

Healthcare PPC keywords should be chosen by intent, not volume. Booking and evaluation searches, such as a specialist in a named area or a procedure cost, tend to produce appointments. Careers, education, self-treatment and medicine searches waste budget and belong on negative lists. Review search terms weekly at first and judge keywords on honoured appointments, not clicks.

Most healthcare PPC keywords lists I review are built from a keyword tool sorted by volume. That is how an orthopaedics campaign ends up paying for “orthopaedic surgeon salary”, “orthopaedics notes PDF” and “knee pain home remedy” alongside the searches that actually book appointments. Medical words are shared by patients, students, job seekers, caregivers and the merely curious, and the ad platform does not know the difference unless you tell it.

This article is about telling it. It covers how I sort keywords by intent, the patterns that tend to book appointments in Indian search, the ones that quietly waste budget, how match types and negatives behave today, and how to judge a keyword on appointments rather than clicks. It is part of the complete guide to Google Ads for doctors in India.

Sort healthcare PPC keywords by intent, not volume

Before building ad groups I sort every candidate keyword into one of four tiers. The tier decides whether it gets budget, what page it lands on and whether it belongs on a negative list.

Booking intent. The searcher wants to see someone: a specialist or procedure with a place, “near me”, “appointment”, “consultation”, “hospital”, “clinic”, “surgeon”, “doctor open now”.
Evaluation intent. The searcher is choosing: “cost”, “price”, “package”, “best”, “second opinion”, insurance and cashless queries. High value, but only with a page that answers the question.
Research intent. The searcher is learning: symptoms, causes, “what is”, “home remedies”, “diet for”. Usually better served by content than by paid clicks.
Non-patient intent. Jobs, salaries, courses, exam preparation, notes, PDFs, images, meanings. This tier goes straight to negatives.

Most budget problems I see are tier 3 and tier 4 terms sitting in campaigns built for tier 1. The fix is structural, not a bid change.

The tiers also tell you where paid search should stop and content should take over. Research-intent searches are real demand, and some of those searchers become patients months later. They are simply cheaper to reach with a clinically reviewed article that ranks than with a paid click that lands on a booking page they are not ready for.

Keywords that tend to book appointments

The patterns below are the high-intent medical keywords that most often turn into booked visits in the accounts I have worked on. They are patterns, not a list to paste, and each should be checked against your own search terms report:

Specialist plus place: “cardiologist in Nagpur”, “gynaecologist Whitefield”, “child specialist near me”. Include the neighbourhood names patients use, not only the city.
Procedure plus provider word: “knee replacement surgeon”, “cataract surgery hospital”, “hernia operation doctor”.
Appointment language: “book appointment dermatologist”, “ENT consultation today”, “Sunday OPD”.
Cost and package queries: “knee replacement cost in Indore”, “full body checkup price”. Valuable if your page shows a range or package; wasteful if it does not.
Insurance and cashless: “cashless hospital for cataract”, queries naming a TPA or insurer. Strong intent where you are empanelled.
Second opinion: “second opinion for spine surgery”, “cancer second opinion”. A patient already in a decision, covered in second-opinion demand.

Specialty shapes this. Orthopaedic patients often delay and research for a long time before searching with booking intent, a pattern I describe in orthopaedics online: marketing to patients who delay. Oncology searches are dominated by caregivers and second opinions, which is why cancer patients search differently. Paediatrics is almost entirely parents searching on a phone, often late in the evening.

Keywords that waste budget

These are the terms I add to negative lists before launch, then extend from the search terms report:

Careers and education: jobs, vacancy, salary, course, fellowship, NEET, MBBS, internship, notes, PDF, PPT, syllabus.
Information seekers: meaning, definition, “in Hindi”, Wikipedia, images, diagram, symptoms (unless you have a deliberate content campaign).
Self-treatment: home remedies, exercises, diet, yoga, “without surgery” (check this one carefully for your specialty).
Medicines and drug names: wasteful and, in India, a policy risk too, as covered in Google Ads healthcare policy.
Wrong service or setting: veterinary, ayurvedic or homeopathic terms if you do not offer them, government hospital names, services you do not provide.
Other providers’ names: searches for a specific competitor hospital rarely produce patients who were open to you.

Two words need judgement rather than a reflex. “Free” is waste for most hospitals, but not if you run free screening camps with their own campaign. Government scheme names are waste if you are not empanelled and high intent if you are. Decide per unit, not once for the whole group.

My starting lists by specialty are in the specialty negative keyword lists, organised so they can go straight into shared lists.

Map every keyword to one page and one ad group

A keyword is only as good as the page it lands on. “Knee replacement cost in Indore” sent to a generic orthopaedics page will underperform the same keyword sent to a page with the package, what it includes and how cashless works. So once keywords are sorted by intent, I map each one to the ad group and page that answers it most directly.

The mapping tends to follow a simple pattern:

Specialist plus place goes to the department or doctor page for that unit.
Procedure plus provider word goes to the procedure page with the team who performs it.
Cost and package queries go to a page that shows price guidance.
Insurance and cashless queries go to a page that lists empanelments and explains the process.
Second opinion queries go to a page that explains how to send reports and what happens next.

If a keyword has no page that answers it, either build the page or leave the keyword out. How this mapping shapes campaigns and ad groups is covered in campaign structure by specialty and procedure, and the page itself in landing pages for doctor consultation ads.

How match types behave now

Match types no longer work the way many agency decks still describe them. Google’s match types help page is explicit that all three match on meaning:

Broad match can show ads on searches related to the keyword, including ones that do not contain its words. It uses signals such as the landing page and other keywords in the ad group.
Phrase match shows ads on searches that include the meaning of the keyword.
Exact match shows ads on searches with the same meaning or intent.

In healthcare that breadth cuts both ways. Broad match finds real patients who phrase things unusually, and it also finds students and job seekers. My default is phrase and exact for launch, with broad match introduced only in ad groups with reliable conversion tracking, strong negatives and a bidding strategy that optimises for appointments. AI Max for Search campaigns pushes further in the same direction, adding broad and keywordless matching on top of your keywords, so it needs the same preconditions.

Negative keywords for hospitals: the details that matter

Negatives behave differently from positive keywords, and the negative keywords help page spells out how. The details that catch hospital accounts:

Negatives do not match close variants. Excluding “job” does not exclude “jobs”. Add plurals and synonyms yourself. Google does handle misspellings.
Negative broad blocks a search only if it contains all the negative’s words, in any order.
Negative phrase blocks searches containing those words in that order.
Negative exact blocks only that exact search.
Account-level negatives apply across eligible search inventory, including Performance Max, which is where universal exclusions such as careers and courses belong.

The common mistake in the other direction is over-negating. I have seen “cost” added as a negative because a few price shoppers annoyed the call centre, which removed some of the most decisive searches in the account. Every negative should have a reason written against it.

Mining the search terms report every week

The search terms report shows the actual searches that triggered your ads. It is where you learn what you are paying for, and in the first two months it deserves a weekly slot in someone’s calendar. Some low-volume terms are withheld for privacy, so it never shows everything, but it shows enough.

The routine I ask teams to follow:

Sort by cost, not by clicks, and read the top of the list first.
Tag each term with its intent tier.
Add tier 4 terms to the right shared negative list, with plurals and synonyms.
Move good tier 1 and tier 2 terms into the ad group whose page answers them, as keywords in their own right.
Note new patterns, such as a neighbourhood name or a Hinglish phrasing, and check whether ads and pages speak to them.
Record what changed and why, so the next person understands the list.

A pre-launch keyword checklist

Before any new campaign goes live, I run the keyword set through a short checklist:

Every keyword is tagged with an intent tier, and tier 3 and 4 terms are either removed or deliberately placed in a separate content campaign.
Every keyword maps to one ad group and one landing page that answers it.
No drug names, experimental treatment terms or other policy-sensitive words are present.
Shared negative lists for careers, education, medicines and information-seeking are attached, with plurals and synonyms.
Location terms match the catchment the campaign actually targets.
Terms that depend on empanelment or a specific doctor are flagged so they can be paused when that changes.

It takes an hour for a specialty and saves weeks of paying to learn the same lessons from the search terms report.

Regional language and tier-2 search

Keyword tools under-represent how many Indians actually search. In tier-2 and tier-3 cities especially, patients type Hindi or regional words in Roman script (“ghutne ka doctor”, “pathri ka operation”), mix English and local terms, and use colloquial names for conditions and procedures. These queries often have low reported volume and real booking intent.

Spelling varies too. The same procedure can be typed several ways, and doctors’ own terminology rarely matches what patients type. Google handles misspellings for negatives, but for positive keywords the practical answer is to let phrase or broad match catch variants once tracking is reliable, and to watch the search terms report for the phrasings that matter.

The search terms report is the most reliable source for them. When they appear and convert, give them their own keywords and, where volume allows, ads and pages that speak the same way. Ask the front-desk and contact centre team too; they hear exactly how patients describe what they need.

Judge keywords on appointments, not clicks

Click-through rate and cost per click tell you whether a keyword attracts attention. They say nothing about whether it produces patients. The keywords I cut most often look healthy on click metrics: plenty of clicks, reasonable cost, and almost no honoured appointments when matched against the register.

The quarterly review I run is simple. For each keyword theme, list spend, enquiries, appointments booked, appointments honoured and, where the data allows, patients treated. Themes that produce enquiries but few honoured visits usually point to a mismatch: the wrong page, the wrong location, or a query that looks like booking intent but is not. Themes with few enquiries but strong conversion to visits often deserve more budget, not less.

That requires tracking that follows the enquiry to the visit, at least for calls and forms. The setup is covered in call tracking and offline conversion import for hospitals, and the discipline of judging spend by outcome is the core of performance marketing for hospitals: the spend traps. Once that data exists, a quarterly keyword review by honoured appointments does more for efficiency than any amount of bid tuning.

Questions people ask

What are healthcare PPC keywords?

They are the search terms a hospital, clinic or doctor bids on in Google Ads so that ads appear when people search for their services. The useful ones signal booking or evaluation intent, such as a specialist in a named area, a procedure with a provider word, or a cost query. The wasteful ones share medical words but come from students, job seekers or people researching symptoms with no plan to visit.

As a unit head, how do I know if our keywords are wasting money?

Ask for the top search terms by spend for the last month, not the keyword list. If you see jobs, courses, notes, home remedies, medicine names or other hospitals’ names near the top, money is leaking. Then ask how many ad enquiries became honoured appointments, by keyword theme. Themes with healthy clicks and few visits are candidates for cuts or better landing pages.

Should we bid on symptom keywords?

Usually not in booking campaigns. Symptom searchers are mostly researching and are better served by clinically reviewed content that ranks organically. There are exceptions where a symptom strongly implies a specific specialist visit, but test those in a separate ad group with its own page and judge them on appointments. Never let symptom terms creep into consultation ad groups through broad match unnoticed.

Are cost and price searches worth paying for?

Often yes, because the searcher is close to a decision. They only pay off if the landing page answers the question with a package, a price range or a clear explanation of what drives cost. Sending cost searches to a page with no price information produces bounces and frustrated calls. If your hospital cannot publish any price guidance, consider excluding cost terms until it can.

Which match type should a new account use?

I start with phrase and exact match, because they keep the early search terms closer to what you intended while you build negatives and verify tracking. Broad match can work well later, but it matches on meaning and relies on good conversion data and negative lists to stay on target. Introduce it ad group by ad group, and compare results on honoured appointments, not on reported conversions alone.

How often should negative keyword lists be updated?

Weekly for the first two months of any new campaign, then fortnightly or monthly once the search terms settle. New patterns appear whenever you add a service, a location or a broader match type. Keep shared lists by theme, such as careers, education and medicines, so updates apply everywhere at once, and note the reason for each addition so future reviewers do not undo it.

As an agency, what is the most common negative keyword mistake?

Assuming a negative covers its variants. Negative keywords do not match close variants, so excluding job leaves jobs untouched, and plurals and synonyms must be added separately. The second most common mistake is over-negating valuable terms like cost or best because a few calls were unhelpful. Every negative should have a written reason and be reviewed when services or empanelments change.

Do we need keywords in Hindi or regional languages?

Look at the search terms report before deciding. Many patients, particularly in smaller cities, type Hindi or regional words in Roman script or mix them with English. Those searches often show little volume in keyword tools but carry real booking intent. Add them as keywords when they appear and convert, and where volume justifies it, match them with ad copy and landing pages written the same way.

Should we bid on competitor hospital names?

I advise against it for hospitals and doctors. People searching a specific hospital usually want that hospital, so the clicks rarely become patients. Using another provider’s name in ad text also invites trademark complaints. It sits poorly with the professional tone patients expect from healthcare. Put that budget into your own specialty and brand campaigns, where intent is on your side.

How do government scheme searches fit in?

It depends on empanelment. If a unit is empanelled under a scheme and has the capacity and processes to serve those patients, scheme-related searches can be high intent and deserve their own ad group and page. If not, they create enquiries you cannot serve and should be negatives. Decide unit by unit, and revisit when empanelment status changes.

How long before we know which keywords work?

Search term quality is visible within the first few weeks, which is when to clean negatives. Knowing which keywords produce honoured appointments takes longer, usually two to three months, because enough enquiries have to flow through to visits and be matched. Low-volume specialties take longer still. Resist cutting keywords on click metrics alone before that appointment data exists.

As a doctor, what can I contribute to keyword research?

The words your patients actually use. Tell the marketing team how patients describe their problem at the first consultation, which terms they confuse, which procedures they ask about by colloquial names, and which enquiries you do not want. That list is often more useful than a keyword tool. Also flag any procedure or treatment name that might raise policy or clinical accuracy concerns.

How do keywords scale across multiple units?

Keep a central keyword and negative framework by specialty, then add local layers: neighbourhood names, local language phrasings, empanelment-dependent terms and unit-specific services. Shared negative lists for universal waste, such as careers and education, should apply across the account. Review search terms by unit, because patterns differ between a metro campus and a tier-2 unit even within the same specialty.

Read my takes first in Google Search