Google Ads for Telehealth: Certification, Keyword Strategy, and Why Search Is Your Cheapest Patients
Most telehealth brands are built on Meta and treat Google as an afterthought - usually a neglected brand campaign and a half-configured Performance Max someone turned on during a slow quarter. That’s backwards for one structural reason: a person typing “online doctor for insomnia” has already done your persuasion for you. On Meta you create demand; on search you collect it, and collected demand converts at rates and CACs that cold social traffic rarely touches.
Across partner accounts, high-intent non-brand search is consistently among the cheapest patient acquisition available - when the account is certified, structured, and fenced correctly. Each of those three is where telehealth brands get hurt. Here’s the playbook.
What certification does Google require before you can advertise?
Google gates healthcare advertising hard, and the gate depends on what you actually offer:
| What you offer | What Google requires |
|---|---|
| Telehealth consultations (no Rx fulfillment) | Healthcare provider verification for telemedicine advertisers |
| Prescribing / online pharmacy dispensing | LegitScript certification + Google’s online pharmacy verification |
| Addiction treatment services | LegitScript addiction-treatment certification |
| Supplements only | No certification, but strict policy limits on claims and ingredients |
Three things founders consistently get wrong here:
- Timeline. LegitScript certification takes weeks to months, has real documentation requirements (licensing, fulfillment partners, policies), and carries fees. If Rx telehealth is on your roadmap, start certification before you need the channel, not after.
- Scope. Certification permits you to advertise; it doesn’t exempt you from policy. Drug names in ad copy, unapproved keyword-to-lander combinations, and claim violations still trip disapprovals that accumulate into account-level trust damage - same dynamic we cover on the Meta side in our GLP-1 compliance playbook.
- The uncertified workaround. Without certification you can often still advertise the service - consultations, provider access - as long as ads and landers stay off the medication. That’s a legitimate interim strategy; pretending to be a consultation-only service while your lander sells prescriptions is how accounts die.
Should you capture demand or create it on Google?
Capture first - fund demand creation only after capture is saturated. Both exist on Google, and confusing them wrecks budgets. The mental model:
- Demand capture - search terms where the patient already wants a solution: condition + treatment (“online TRT clinic”), competitor names, “near me”/“online” service queries, and your brand. Highest intent, finite volume. This is where you start and where the cheap patients live.
- Demand creation - YouTube, Display, Discover. Real reach, but it behaves like social: it needs creative volume, longer windows, and it will not show search-like CACs. Fund it after capture is saturated, not instead.
The practical sequencing we run at AdBoost Health: own brand first (cheap insurance, and competitors will bid on you), then condition/treatment non-brand, then competitor terms once your lander can win the comparison, then YouTube when search impression share says you’ve bought all the intent that exists. What “good” looks like at each stage varies enormously by vertical - anchor expectations with our telehealth CAC benchmarks before judging a four-week-old campaign.
How should you structure keywords and negatives?
Structure campaigns by intent tier, keep match types on a leash, and build the negative list before launch - because health search has a property most categories don’t: enormous informational query volume that looks transactional to broad match. “Semaglutide side effects,” “does TRT cause hair loss,” “insomnia in pregnancy” - a loosely fenced account will happily spend your budget educating people who are researching, not buying, or who can’t be served at all.
Structure that holds up:
- Campaigns by intent tier, not just by condition. Brand / non-brand core / competitor as separate campaigns with separate budgets, so high-intent terms never share a budget cap with experiments.
- Match types on a leash. Exact and phrase carry the core; broad match only paired with strong conversion signal and daily query review in the first weeks. Broad match in health without negatives is a donation to Google.
- A negative list you build before launch, not after the bill. The standing blocks for telehealth: information-intent modifiers (“side effects,” “reviews,” “vs,” “reddit,” “what is”), job seekers, “free”/“cheapest” bargain hunters, ineligible geographies and populations you can’t treat, and - critically - safety-sensitive queries (overdose, interactions, emergency symptoms) where showing an ad is both useless and reputationally radioactive.
- Weekly search-term hygiene. In health, the query landscape shifts with news cycles and drug trends; last quarter’s negative list has holes in it today.
Why does Performance Max need a leash in health?
PMax needs a leash in health because its automation hides queries, cannibalizes brand, and drifts placements. It’s Google’s push-button answer, and in most verticals it’s fine - in telehealth it needs adult supervision:
- Query opacity. PMax hides most search terms, which means you can’t fully see whether you’re showing against safety-sensitive or non-compliant queries - a compliance posture problem, not just an optimization one.
- Brand cannibalization. Left unfenced, PMax loves to “win” your own brand searches and report them as performance. Apply brand exclusions or you’re paying a platform tax on demand you already owned.
- Placement drift. Health creative auto-assembled onto random Display inventory is how disclaimers get cropped and claims get re-contextualized.
- Lead quality blindness. PMax optimizes to the conversion you feed it; feed it form-fills and it will find you infinite unqualified form-fills. It only behaves when fed a deeper event - booked visit or qualified intake - via proper conversion imports.
Our default: prove the account on tightly structured Search first, add PMax later as a scaling layer with brand exclusions, filtered conversion goals, and constant creative review - never as the foundation.
How do you make search and social work together?
The channels compound: Meta creates the demand, and thirty seconds later that same patient is on Google typing your brand or your category. Brands that measure them in isolation systematically over-credit search and under-credit social - which is an attribution problem before it’s a media problem, and why we treat fixing telehealth attribution as a prerequisite to scaling either channel.
Run them as one system: social creative volume feeding demand, search structured to catch it, negatives protecting the budget, certification keeping the whole thing alive. Setup for that system takes us about 5 days on a new partner account, and it’s the pattern behind most of the $153M+ in tracked revenue across the 71+ founders we’ve worked with.
If your Google account is either dormant or quietly burning budget on the wrong queries, book a free strategy call - 30 minutes, we audit the account live, and you get a written plan whether or not we ever work together.
