If your phone is how your business gets work, there is a Google Ads change coming that nobody sent you a letter about.
The call-only ad is being switched off. It is the ad type that shows a phone number instead of a website link, runs on phones only, and dials you the moment someone taps it. For a plumber at eleven at night, a tow operator, an emergency dentist, a bail bondsman, it has been the single most efficient thing in the account for a decade. No website needed. No form. Somebody with a problem taps once and your phone rings.
Google announced its retirement in October 2025. Most owners running these ads still have not heard.
Two Dates, and You Are Already Past the First One
The retirement happens in two steps, and Google has published both.
February 2026: you can no longer create a new call-only ad. The option has already been removed from the interface. If you go looking for it in a search campaign today, it is not there. That date has passed.
February 2027: existing call-only ads stop receiving impressions. Anything still running keeps running until then, and then stops.
So you are in the middle of the window right now, with roughly six months of it left. If you have call-only ads live in an account, they are still serving, and that is exactly what makes this easy to ignore. Nothing looks broken. The leads still come in. The deadline is on a support page you have never read.
There is one more wrinkle worth knowing before you plan around this. Because you cannot create new call-only ads any more, you also cannot rebuild one if something goes wrong with an existing one. If a call-only ad gets disapproved on a policy issue today, fixing it is not an option. That ad is gone, and the campaign is down until you build its replacement. The safety net came out in February, not next year.
The replacement Google points to is a responsive search ad with a call asset attached. That sounds like a straight swap. It is not, and the difference is the whole point of this article.
The Real Deadline Is Earlier Than the Stated One
February 2027 is when the lights go out. It is not when the performance goes.
Here is the pattern anyone who has run these accounts for a few years will recognize. When Google decides a format is finished, it rarely stays healthy until the switch-off date. Impressions thin out. The format stops winning auctions it used to win. The system quietly favors whatever it is steering you toward, and by the time the official date arrives the thing has been dying for a year.
Chris at Downey Marketing, who runs Google Ads for local lead generation, has been watching exactly that happen with call-only for the past two years: less traction in search, with responsive search ads increasingly favored. That matches what we see. I want to be straight about the status of that claim, because it matters. Google has not published anything saying impressions will taper, and I cannot prove they will. It is a pattern, not a policy.
But it is the pattern I would plan around. Treat the deadline as the point by which the format is worthless, not the point at which it stops working.
What You Are Actually Losing
Before rebuilding anything, it is worth being honest about what made the call-only ad good, because the replacement does not do all of it.
A call-only ad did one job with total clarity. It appeared on phones only. It showed your phone number as the headline. Tapping it opened the dialer with your number already in it. There was no website in the path, no page to load, no form to fill, no second decision to make. You did not even need a website, only a page Google could check that carried the same number.
That is a shockingly short distance between a person's problem and your phone ringing. For an emergency trade, short distance is the entire business. Somebody standing in four inches of water is not browsing. They are going to call the first plausible number they see, and the format was built for precisely that moment.
None of the replacements are as short. That is the honest starting position, and it is why this deserves an afternoon of your attention rather than a shrug.
A Call Asset Is Not a Call-Only Ad
This is the part that gets glossed over everywhere, including in Google's own framing, and it is the thing you actually need to understand.
Call assets used to be called call extensions. They attach a phone number to a normal responsive search ad. When one shows, the searcher sees your ad with a tappable number next to it, and a share of people tap the number instead of the headline.
Here is the difference nobody states plainly. A call-only ad always showed your phone number, because the number was the ad. A call asset shows when Google judges it worth showing for that particular search, on that device, for that person. Assets are eligible to appear, not guaranteed to appear.
You have moved from a guarantee to a maybe.
That is not a reason to panic, and it is not a reason to distrust call assets, which work well. It is a reason to stop thinking of this as a swap and start thinking of it as a rebuild. Once the number is only sometimes on the ad, the rest of the path has to carry the calls the ad no longer guarantees. Which means the landing page you never needed is now load-bearing.
Rebuilding It: Four Moves
You cannot recreate a call-only ad. You can recreate a call-only campaign. Four changes, all of which take one sitting.
One: build the responsive search ad. In the ad group, add a responsive search ad in place of the call-only ad. You get up to 15 headlines and 4 descriptions, and Google assembles combinations from them. Write them for the emergency, not for your company history. "24 Hour Emergency Plumber," "We Answer at 2AM," "Burst Pipe? Call Now." Google will find the combinations that win.
Two: add the call asset, and match the number. Under assets, add a call asset with the number you want ringing. The number on the asset must match the number on the page you send traffic to. This trips people constantly. If your ad says one number and your landing page shows another, expect trouble with approval and with the searcher, who now has two numbers and no idea which is real.
Three: force it back onto phones. A call-only ad was mobile-only by definition. A responsive search ad is not. Go to the campaign's device settings and set the bid adjustment for computers to -100% and tablets to -100%. That returns you to phones only, which is what you wanted, and stops you paying for desktop clicks that were never going to become calls.
Four: give it somewhere to land. Every click now goes to a page. This is the new work, and the next section is about it.
Do those four and you have a call-focused campaign again. Different machinery, same intent.
The Landing Page You Never Used to Need
Call-only never sent anyone to a website. That was the appeal, and it is why plenty of businesses running these ads have a website that is old, slow, or barely exists.
Now every click that does not tap the call asset lands somewhere. If that somewhere is your homepage, you will feel it immediately, because a homepage asks a panicking person to make choices.
The page has to do one thing: turn a mobile visitor into a phone call, fast.
- Mobile first, genuinely. Not a desktop site that shrinks. Thumb-sized tap targets, a number that dials on tap, and it has to load on a phone on cellular data in a parking lot.
- No form. This is the one people argue with, and they are wrong for this campaign. If the objective is a call, a form is a competing option that gives the visitor a slower path and gives you a worse lead. Take it off.
- The number is the page. Big, tappable, visible without scrolling, repeated as they scroll. Not in the header at 14 pixels.
- Answer the emergency question above everything else. Are you open now, do you cover my area, how fast can you get here. That is what they are checking before they dial.
- One call to action, repeated. Call. Nothing else.
If you were running call-only because you did not want to build a website, the honest news is that you now need one page. One. It does not need a blog or an about section.
Your Call Reporting Does Not Disappear
Worth saying plainly, because it is the most common worry: moving to call assets does not cost you your call data.
Google's call reporting continues to work with call assets. You can still see calls generated from your ads, their duration, and use call duration as a conversion, which is the setting that matters most for a service business. A 45-second call and a 4-second misdial should not count the same, and they do not have to.
If anything, this migration is a decent moment to check that your call conversion is actually set up sensibly. Plenty of accounts we look at count every call as a conversion regardless of length, which teaches Smart Bidding to chase wrong numbers. If you are rebuilding the campaign anyway, set a sane duration threshold while you are in there. Thirty seconds is the usual starting point, because a voicemail is still a real lead.
Some of You Should Not Rebuild It at All
A forced migration is a decent moment to ask whether the thing you are migrating was right.
Rebuild it if your business genuinely closes on the phone, your team answers reliably, and your work is urgent. Emergency trades, medical and dental, legal intake, locksmiths, towing. For these, the phone is not a preference, it is the product, and you should rebuild it properly and quickly.
Reconsider if you ended up on call-only mainly because it was the simplest thing to launch. Some businesses have been paying for calls they were never well set up to answer. If a meaningful share of your calls go to voicemail during business hours, more calls is not your problem. If your service is considered rather than urgent, a good page and a real offer may beat a rushed phone call.
And if you have never tested it, this is your chance. You are being made to build the page and the responsive search ad regardless. Once both exist, you can find out honestly whether calls or forms produce better customers for you, rather than assuming.
What to Do This Week
Three steps, in order.
Find out whether you are affected. Open the account, filter your ads by type, and see whether any call-only ads are still live. Plenty of owners are not running them and can stop reading. If you are, note which campaigns.
Rebuild one campaign, not all of them. Take the highest-spending campaign that depends on call-only, do the four moves, build the one landing page. Run it alongside the existing call-only ads rather than deleting them, since they still serve until 2027. Watch it for a couple of weeks.
Then migrate the rest, and delete the old ads deliberately. Once the rebuilt version is producing calls at a cost you recognize, move the others and remove the call-only ads yourself. Better to switch them off on a day you chose than to discover in 2027 that they quietly stopped working months ago.
We do this transition on client accounts most weeks at the moment, and the pattern is consistent: the businesses that struggle are not the ones with bad ads, they are the ones with nowhere good to send the click. If you would rather not work out the page and the campaign yourself, we run Google Ads for local service businesses and are happy to look at what you have. Either way, go and check whether you have call-only ads live. It takes two minutes, and the answer decides whether the next six months matter to you.
