The 90 second rule for phone bookings Google can't see

An online booking is easy to count. When it's set up properly, the moment someone completes the booking, a signal fires and Google records a conversion. Clean. Done.

A phone call is a different animal. Google can tell you the ad was clicked and the call connected. After that it goes dark. It doesn't sit on the line. It doesn't hear 'great, see you Thursday at 10.' For a profession where the receptionist answering the phone is still the front door, that's a huge blind spot.

So the question becomes: do you just throw those calls away because you can't track them to completion, or do you make up a number that flatters the account? Neither is good enough when you're deciding whether to spend more money next month.

The first wrong way is to ignore phone calls entirely. The report only counts online bookings, the phone bookings vanish, and a perfectly healthy account looks like it's failing. You spent $1,000, the report shows 4 online bookings, and you quietly decide Google Ads doesn't work for your clinic. Meanwhile 6 people called and booked, and nobody counted them.

The second wrong way is to guess generously. Every call gets treated as a booking, the cost per new patient looks incredible, and you scale up a campaign that isn't actually performing the way the numbers claim. That's the same lie as a 148 percent conversion rate, just dressed up nicely.

Both versions leave you managing a budget against a number you can't trust. And if you don't trust the number, you can't sensibly move the budget up or down, because you genuinely don't know whether it's working.

Here's the proxy I actually use. I report two things: how many people booked online, and how many phone calls lasted longer than 90 seconds. In my words to clients: "I report back how many phone calls did you have longer than 90 seconds, and how many people booked online."

The logic is blunt. A call under 90 seconds is almost never a booking. It's a wrong number, a 'do you do this, no, ok bye,' a quick question, an existing patient moving an appointment. "Nothing less than 90 seconds, I'm like, they just didn't book." Past 90 seconds, something real is happening. "As long as it's longer than 90 seconds, there's a chance they booked a new client."

I want to be honest about what this is. The 90 seconds is an illustrative threshold, not a law of physics. Some clinics book in 70 seconds, some take three minutes. But as a line in the sand it works remarkably well, and it gives you a clean, defensible count of calls that probably mattered. From there you apply a sensible booking rate. I default to around 50 percent, meaning if you had 10 qualifying calls you estimate 5 of them turned into a new patient. That rate is also illustrative. If you know your phone converts better or worse, use your own number.

This is easier to feel than to read about, so I built a small calculator for it. Put in your ad spend for the period, your online bookings, your count of phone calls longer than 90 seconds, and your assumed booking rate.

It gives you back an honest total of new patients (online plus estimated phone) and a true cost per booking that includes the phone. Watch what happens to that cost per booking when the phone bookings stop being invisible.

The point of the 90 second rule isn't precision. It's honesty. You're not pretending to know exactly which calls booked. You're stating an assumption out loud, applying it consistently, and putting a real number in the report you can argue with. That's the difference between a metric and a fairy tale.

The 90 second rule fixes the phone side. Two more habits keep the whole picture honest.

First, get your online bookings tracked as close to 100 percent as you can. When the booking is embedded on your own page or routed through your practice management portal, tracking accuracy sits around 100 percent. When you link out to a separate booking page, accuracy drops to around 90 percent, because a chunk of the journey happens somewhere Google can't watch. Same bookings either way. The difference is purely your ability to measure them.

Second, count One, not Every. If conversions are set to count every conversion, the same person becomes a conversion twice. A returning patient who Googles you to call or shift an appointment hits your ad, you pay for it, and it logs again. Counting One stops you double counting the same caller and double paying for traffic you'd have got anyway. Pair that with the 90 second rule and you've got a phone number that means something.

A clinic owner doesn't need a 12 page report. You need three numbers: how many new patients, what you spent, and what you paid per new patient. The phone is part of all three. Leave it out and every one of those numbers is wrong.

Get the phone counted honestly and the account stops being a mystery. You stop switching off campaigns that were quietly working, and you stop scaling campaigns that only looked good because every call got counted as gold.