Google Ads
Your care plan patient is worth more than your one off. Tell Google that.
Why every booking counting as one identical conversion quietly costs you your best patients.
By Pete Flynn · 29 June 2026 · 8 min read
Here is a sum most clinic owners have never actually sat down and done. Take the patient who comes in once for a tweaked back, gets sorted in a visit or two, and you never see again. Now take the patient who starts a full care plan, or the funded patient who is with you for the better part of a year. One of them might be worth a couple of hundred dollars. The other might be worth a few thousand. Same clinic, same ad, wildly different value walking through your door.
What the bidding can see
Three very different patients. One number, or three.
What most accounts send
Every booking counts as one identical conversion.
One off
Care plan
Funded
Same height, same value. The bidding has no reason to prefer the patient worth thousands over the one worth a couple of hundred, so it chases whatever is cheapest to win.
What a valued account sends
Each booking type carries its own value.
One off
Care plan
Funded
Now the gap is visible. The bidding learns to fight harder for the clicks that turn into the patients who actually pay your rent.
The trap: value rules cannot read patient type
What Google can see at auction
Location, audience, device. That is the whole list conversion value rules can adjust on.
What it cannot see
Whether the lead is acute, a care plan or funded. So the only way to value patients differently is separate booking actions, each with its own value.
Every booking is not worth the same, and your account knows none of it
In almost every clinic Google Ads account I open, both of those patients are counted exactly the same way. One booking, one conversion, job done. The acute one off and the year long funded patient land in the account as two identical green ticks.
Google's job is to spend your budget as fast as it can and your job is to get as many good new clients as you can, two slightly different goals, so when it goes hunting for more bookings it has no idea that one of those ticks is worth ten of the other. It just buys whatever is cheapest to win. And the cheapest to win is almost never your best patient. The valuable ones are usually the harder, more competitive, more expensive clicks, so a system told they are all equal will quietly steer away from them.
I run Google Ads for over 120 Australian clinics, and to be honest this is one of the most common bits of money I see left sitting on the table. The fix itself is not that complicated. The catch is that the two tools most people reach for to do it are the wrong ones.
If every booking counts as one identical conversion, the bidding has no idea your best patient exists. It just buys you the cheapest one.
The two settings everyone reaches for, and why they both miss
When owners hear this, they go hunting for a setting. There are two obvious candidates, and I want to save you the detour, because I have taken it myself.
The first one is conversion value rules. On paper they sound perfect for this, but they only let you adjust a conversion's value by the handful of things Google can actually see at the moment of the auction. Location, audience, device. That is the list. None of them tell Google whether the person clicking will turn into a one off or a year long funded patient, because nobody knows that yet at the click. So value rules are genuinely useful for saying a booking in your best catchment suburb is worth a bit more. They cannot do acute versus care plan versus funded. Wrong tool.
The second is the bidding strategy itself, Target ROAS, which does chase value. The catch there is volume. To behave itself, Target ROAS really wants somewhere north of 50 value carrying conversions a month, with accurate values attached. Most clinics on a sensible budget are doing 5 to 25. Below about 30 a month the signal is too thin, and a data starved Target ROAS will either starve your everyday bookings or thrash around chasing the occasional big one. At that size a plain Maximise Conversions or a Target CPA usually beats it. So this is the right idea at the wrong scale, and if it was my account I would leave Target ROAS alone until the volume is there.
Conversion value rules
- Adjust value by location, audience or device
- Handy for a higher value suburb
- Cannot tell an acute visit from a care plan from a funded patient
- The wrong tool for valuing patient types
Separate booking actions
- One conversion action per booking type
- Each one carries its own default value
- Works from day one, no volume threshold to clear
- Captures the real gap between your patients
The way that actually works on a small budget
Here is the move. Instead of one Book Now conversion that lumps everything together, you set up a small handful, one per kind of booking. Book, one off. Book, care plan. Book, funded. Each one gets its own default value, and now the bidding can finally see the difference.
You do not need a perfect number for those values. Take what that patient is genuinely worth to you over their time on your books, then multiply it by how often a booking of that kind actually shows up and becomes that patient. That gives you a sensible proxy value to put on the action. Imperfect values applied consistently beat no values at all. That is the whole game.
One thing to keep honest: scale the numbers sensibly against your clicks. If your clicks cost five to fifteen dollars and you tell Google a booking is worth three thousand while another is worth one, you can send it chasing the rare big fish and starving your everyday flow. Keep the values on a coherent scale. The calculator below does that arithmetic for you.
Setting it up, start to finish
Step 1
Split the booking action.
Replace the single catch all conversion with one per booking type. One off, care plan, funded. Same booking button, three labelled outcomes.
Step 2
Put a value on each.
Use the proxy method: what the patient is worth times how often that booking becomes that patient. Static defaults are completely fine to start.
Step 3
Feed real values back later.
Once your practice software knows the true value, you can feed it back through enhanced conversions for leads or an offline import. Useful, but not where you start.
Your patient value mix
Stop telling Google every booking is worth the same.
Put in what each kind of patient is actually worth to you over their time on your books, and roughly how often a booking of that kind turns into that patient. The calculator shows how many times more your best patient is worth, and the value to set on three separate booking conversion actions so the bidding learns to chase the ones that pay.
An acute visit or two, then gone
A full plan of sessions through to discharge
NDIS, plan managed or compensable, over a year
Of bookings that show and become a real patient
Your most valuable patient is worth
15.9× your one off
$3,500 against $220. If every booking counts as one identical conversion, the bidding has no idea this gap exists, and it will happily buy you ten of the cheap one and none of the valuable one.
The value to set on each booking action
Book, one off
$132
Conversion value
Book, care plan
$840
Conversion value
Book, funded
$2,100
Conversion value
How to actually use these
Set these as the default value on three separate booking conversion actions, one per booking type. Do not try to do this with conversion value rules. Those only adjust by location, audience and device, and none of them can tell Google whether a booking is acute or a funded plan. Imperfect values applied consistently beat no values at all.
These are proxy values, deliberately gross revenue based. The point is the relative gap between them, not the cents. Get clean booking completion tracking working first, because values layered on broken conversions just teach the bidding to chase noise.
Your best patient is worth 15.9 times your one off, and most accounts tell Google none of it.
The first thing to check is whether your account is even counting real bookings, let alone valuing them differently.
Do this before any of it: count real bookings
Everything above assumes your account is counting real bookings in the first place. For a clinic on five hundred to three thousand a month, just getting clean booking tracking that counts each real booking once, and once only, is a bigger win than any clever value setup.
Values layered on broken conversions just teach the bidding to chase noise. If your account is counting booking button clicks, or contact page views, or get directions taps as bookings, then weighting them by patient type is putting a smarter steering wheel on a car with no engine. Fix the signal first. Then make it smart.
The one migration line you actually need to know
There is a change doing the rounds at the moment that sounds scarier than it is. From the 15th of June 2026, Google is retiring an old programmatic way of uploading offline conversions for new and inactive setups, in favour of its newer Data Manager API.
If your clinic tracks bookings through Google Tag Manager or the standard tag, or you upload the odd conversion by hand, nothing breaks. This is a plumbing change for developers and agencies, not a clinic owner emergency, and most of the panicked posts about it overstate the disruption.
The only thing worth doing is asking whoever runs your tracking one question. Are we allowlisted, or already moved across to the Data Manager API. If you get a blank look back, that tells you something useful about who is actually minding your account.
Not sure your account values anything correctly?
We audit clinic Google Ads accounts and show you exactly what they are counting.
We start with whether the account is counting real bookings at all, then whether it knows a care plan from a one off. The maths either stacks up or it doesn't, and you will see which.
Audit my Google Ads accountCommon questions