This post is for hospital marketing teams that still run call-only ads, or that inherited an account where the phone number is the main conversion. It answers a practical question: now that Google is retiring the standalone call ad format, which of the remaining ways to generate phone enquiries is worth keeping, and how should each be measured?
Phone calls are not a legacy channel for hospitals. Emergency departments, dialysis units, maternity wards and appointment desks all take calls, and many patients in India prefer to speak to a person before they book. So the question is not whether to keep calls as a goal. It is which ad surface should carry them and how you will know the call was worth answering.
The post assumes you already track calls in some form. If you do not, start with call tracking for healthcare, which covers numbers, routing and reporting. This article goes narrower: the format decision and the measurement that follows from it.
What is actually happening to call-only ads
Google announced in October 2025 that it would retire call-only ads and asked advertisers to move phone-led campaigns to responsive search ads with call assets. Its Ads developer team published a note on upgrading call-only ads to responsive search ads with call assets. Trade coverage by PPC Land reports the sequence as follows: creation of new call-only ads was removed from February 2026, and existing ones stop serving in February 2027.
Treat those dates as reported rather than as a promise. Interfaces were reported to change at slightly different times, and Google can adjust timelines. What matters operationally is the direction. Any hospital account that depends on a call-only ad group should have a replacement running well before the final cut-off, so you can compare the two while both are live.
It helps to be precise about what is going away. The standalone ad format, where the headline is a phone number and the click is a call, is being retired. The call asset, formerly the call extension, is not. A call asset attaches a phone number to a normal search ad, so the searcher can read your headlines, then choose to call or to visit the site.
The four ways a hospital can collect a phone call
There are now four routes worth comparing. They differ in who initiates the call, what the searcher has seen first and what Google can measure.
| Route | What the searcher sees | What Google can count |
|---|---|---|
| Call-only ad (retiring) | A number-led ad; the click dials | Calls, with a minimum length you set |
| Call asset on a responsive search ad | A normal ad plus a call button on mobile | Calls from ads, plus separate website clicks |
| Location asset with phone number | Address, map and a call button | Calls from ads when call reporting is on, plus other interactions |
| Click-to-call on your landing page | Your page, then a tappable number | Clicks on the number on mobile; calls only with a website call setup |
Google's help page on phone call conversion tracking describes the same split. Calls from ads covers call-only ads and call or location assets, and uses Google forwarding numbers, so call reporting must be switched on. Calls to a number on your website need the Google tag plus a website call event. Clicks on a number on a mobile site count only the click, not the call itself.
That last point is the most common source of inflated call numbers. A tap on a tel: link is an intent signal. It becomes a call only if the dialler opens and the patient does not hang up before it connects.
Call quality: why the format matters
A call-only ad filters in one direction. The searcher who taps it has decided to phone before seeing your site, so you receive calls from people who prefer speaking to reading. That tends to produce more short, low-intent calls, such as questions about visiting hours, parking or reports. For a hospital, those are real patients, but they are not the same as a booking enquiry.
A call asset on a responsive search ad changes the mix. The searcher can read your headlines and descriptions first, which pre-qualifies them. They can also click through, so some of the intent that would have become a call becomes a form fill or a WhatsApp message instead. If your goal is phone volume, this will feel like a drop. If your goal is qualified enquiries, it is usually the better trade, but you should check this against your own data rather than assume it.
The way to check is a simple comparison over matching weeks, not a sweeping claim. Run the call asset on the same ad groups where call-only ads currently run, keep budgets stable, and compare these measures for each route:
- Calls connected, as a share of call clicks or taps.
- Calls above your qualifying length, by department (use whatever threshold you have today; a flat number is rarely right for every department).
- Calls that were answered, not abandoned or sent to voicemail.
- Appointments booked from those calls, once you can match them.
Hypothetical illustration: a 200-bed hospital in Pune finds that its call-only ad group generates many short calls to the main switchboard, while its call asset on a cardiology ad group generates fewer calls that last longer. Those would be invented numbers, but the pattern shows why connected calls and booked appointments, not call counts, should decide the winner.




