The Dispute Button You’re Looking For No Longer Exists.
Sources: WolfPack Advising 2026, LSA Management in PMax · BG Collective 2026, How Google’s Lead Credit System Works · Google Ads Help, LSA to Performance Max
The first thing most contractors do after migration is go looking for the dispute button to challenge a junk lead. It is not there, and it has not been since 2024. Google now runs the whole thing itself: its machine learning reviews every charged lead within 72 hours and applies a credit automatically if it decides the lead was invalid, usually showing up within 30 days. There is no manual appeal of that decision, and the old credits for out-of-area calls and services you do not offer were quietly discontinued.
Google reviews every charged lead in 72 hours and credits the bad ones itself. Your only input is the survey.
BG Collective 2026, on Google’s automated lead-credit system.
That makes the “Rate this lead” survey the most important two clicks in the account. Marking a bad lead “Very dissatisfied” with a specific reason is what feeds the model and can trigger a credit. It is advisory, not a guarantee, but it is the only say you have left in lead quality, so fill it out on every charged lead. The contractors who treat the survey as busywork are the ones who will watch their effective cost per lead climb and have no idea why.
From the team at DUO
Here is what the migration really changed: Google now runs your bidding on one automated target, and the only way to steer it is the signal you feed it. That is the whole system we have been building in The Brick. We don’t count leads. We count booked jobs, so we feed the campaign booked-job values, watch cost per lead by service line through the learning period, and keep the wheels on while the account settles. If your LSAs just migrated and you would rather not fly it blind, we will run the transition.
Have DUO run your migrated account →Let’s Get Into It
Running LSAs in Google Ads Without Blowing Up Your Cost Per Lead.
The account moved but the old instincts do not apply. Here is the gap, the diagnosis, and the tier-by-tier build to run it right from day one.
The Problem
The account is in a new place, but muscle memory is not. Contractors go hunting for the dispute button that no longer exists, panic-yank the budget on the first slow day of the learning period, and never notice that one blended target CPA is quietly overpaying for the cheap service line to hit the expensive one. Nothing broke. The controls just moved, and a couple of them were removed. Here is where everything went.
The Old LSA Dashboard
Where It Lives Now
Manual per-service bids
One campaign-level target CPA
Weekly budget
Daily budget (weekly divided by 7)
Manual lead disputes
Auto-review plus the “Rate this lead” survey
Standalone LSA dashboard
Google Ads: Goals then Conversions then Leads
The Diagnosis
You are about to make the account worse, not better, if any of these are true:
×
You are still looking for a way to dispute a bad lead by hand.
×
You cut the budget in the first week because a couple of slow days spooked you.
×
You cannot say your cost per lead by service line, so you cannot spot drift.
×
Nobody on your team fills out the lead-rating survey, so the model never hears from you.
The fix is a stack of three phases, tied to the clock: what to do in the first 48 hours, what to do (and not do) through the two-week learning period, and what to run every month once it settles. Start at the top.
The Fix. Three Phases.
Phase 1. First 48 Hours.
Find it, confirm it, set the target.
1 hour. Whoever holds the Google Ads login.
1. Find your campaign in Google Ads. It migrated automatically as a Performance Max pay-per-lead campaign. Confirm it is live, still billing per lead, and still serving on Search and Maps. Your leads are under Goals then Conversions then Leads.
2. Set the campaign-level target CPA. Google now bids to one blended target. Set it off the baseline cost per lead you exported before migration, not a guess. If you never exported it, that number is gone, and you are setting a target blind.
3. Convert your budget to daily. The old weekly number divided by 7. Enter it once and leave it. Do not round up to “make up” for the migration.
Bottom lineLocate it, confirm pay-per-lead, set a target off your real baseline, fix the daily budget. Then stop touching it.
Phase 2. First Two Weeks.
Let it learn. Feed the survey.
10 minutes a day. Whoever handles the leads.
1. Leave the bidding alone. Google asks for about two weeks for performance to stabilize after migration. A slow Tuesday is not a trend. Every time you change the target or budget, you reset the learning and pay for it twice.
2. Rate every charged lead. Open the “Rate this lead” survey and use it on all of them. Mark bad leads “Very dissatisfied” with a specific reason. It is advisory, but it feeds the model and can trigger an automatic credit. It is the only lead-quality lever you have left.
3. Watch Cost / conv. by service line. Check it in Goals then Conversions then Leads. You are looking for one service line whose cost per lead is drifting up under the blended target. Note it. You act on it in Phase 3.
Bottom lineHands off the bidding, survey on every lead, eyes on cost per lead by service line. That is the whole job for two weeks.
Phase 3. Every Month After.
Split what drifts. Reconcile to booked jobs.
2 to 3 hours a month. A named owner.
1. Split out a service line that’s getting expensive. The migration forces one price target across all your services, so a costlier line (roofing, say) can drag the whole account. Pull it into its own campaign with its own target and Google prices it on its own, the per-service control you used to have.
2. Reconcile charged leads to booked jobs. Google will happily optimize to cheap leads. Once a month, match charged leads against jobs that actually booked, by service line, so you are steering on revenue instead of ring volume.
3. Make the survey a standing habit. Build lead rating into your intake so every CSR rates every lead the day it comes in. Over a month it is the difference between the model learning your definition of a good lead and guessing at it.
Bottom lineSplit the drifters, reconcile to booked jobs monthly, make lead rating a habit. This is how the account gets better, not just survives.
Do This Next
☐
Log into Google Ads and find your migrated campaign. Confirm it is live and still pay-per-lead. Leads are under Goals, Conversions, Leads.
☐
Set your campaign-level target CPA off the baseline cost per lead you exported, and convert the weekly budget to daily.
☐
Then do nothing to the bidding for two weeks. Let the learning period run.
☐
Rate every charged lead in the survey. Mark bad ones “Very dissatisfied” with a reason. It is your only credit lever now.
☐
After two weeks, check cost per lead by service line. Split out any line whose cost per lead is climbing into its own campaign, and start reconciling charged leads to booked jobs monthly.
