Google Is Folding Local Services Ads Into Google Ads: The August Migration Checklist

Google is retiring the standalone Local Services Ads platform: the separate dashboard, its bidding controls, and its reports. The program itself is not going away. It is being rebuilt inside Google Ads as a specialized Performance Max campaign type built for pay-per-lead goals.

The rollout is phased. Phase one begins in August for US home and storefront service advertisers, and Google names the trades: roofing, plumbing, HVAC, electrical, appliance repair, house cleaning, lawn care, pest control, and moving. Later this year the migration expands to service-area businesses without a storefront and to accounts with custom bidding or booking configurations. Everyone else, including accounts outside the US, follows in 2027.

Your account administrator receives an email 14 days before migration, a dashboard banner, a reminder at 7 days, and a confirmation when it is done. Two housekeeping notes hide in that sentence: make sure the administrator email belongs to someone who still works at the company, and note that new pay-per-lead campaigns cannot yet be created inside Google Ads itself; new launches still go through the classic Local Services signup.

Manual bidding is being deprecated. Today you can set a maximum cost per lead: a hard wall that nothing gets through. The new system replaces it with an automated target. Those are not the same thing. A maximum is a ceiling; a target is an average, which means individual leads can come in far above it while the system stays exactly on plan.

Picture it as two charts. Under a maximum, every lead price stops at the wall you set; a $200 cap means no lead ever costs $201. Under a target, individual leads can come in at $400 or $600 while the system remains exactly on plan, because the average still lands near $200. The practical response is a habit change: stop watching only the average and review your individual lead prices weekly, so a drifting distribution shows up in week one rather than on the monthly bill.

Weekly budgets become daily budgets. The published math is simple: your historical average weekly budget divided by seven, with monthly spend capped at roughly that daily figure times 30.4. Run it out and the monthly ceiling is essentially unchanged on paper.

The danger is the phrase “historical average weekly budget.” The documentation does not specify whether that means the number you set or the amount you actually spent. Plenty of accounts set a deliberately high weekly budget as a visibility signal and spend a fraction of it. The old platform never chased that ceiling. Performance Max is designed to spend its budget. If the set number is what migrates, an account that quietly underspent every week is suddenly inside an automated system with a big ceiling, no maximum cost per lead, and official guidance telling you to allow up to two weeks for numbers to stabilize before reacting.

The protection costs five minutes: before your migration date, set the weekly budget to a number you would genuinely accept spending. Then the ambiguity cannot hurt you under either interpretation. And apply the two-week patience to lead performance only; check your spend every single day from day one.

The junk-lead fear making the rounds is mostly misplaced. This campaign type has no website forms and no Display placements; you are charged for calls, messages, and bookings routed through Google's own system on Search and Maps. A disconnected number cannot call your phone.

The subtler risk is definitional. The system optimizes to produce leads, and Google decides what counts as one. A solicitor, a job-seeker, a homeowner gathering a fourth estimate, someone eleven months from buying: every one of those is a real call and individually defensible as billable. None of them needs work done now. You will not spot the shift on any single lead; the tell is aggregate. Lead count up, cost per lead flat, close rate falling.

Worth knowing: manual lead disputes were already retired at the end of 2024 in favor of automated credits, and the migration documentation does not describe a credit or dispute process on the new side. The lead feedback form survives, so keep rating every lead, and document everything: dates, recordings, notes. If credits ever get contested, the operators with records are the ones who get made whole.