If you feel like you are getting too many spam LSA leads, you are not imagining it. A large share of raw Local Services Ads leads — third-party estimates put it around 45 percent — are unbookable for one reason or another, and outright spam is a subset of that. Because LSAs charge per lead rather than per click, every junk lead that slips through is real money and real wasted time. The goal is not to chase every dollar back after the fact; it is to reduce the junk at the source and triage the rest fast.
This guide breaks down the types of junk leads, what Google's auto-credit system will and won't cover, how to use the "Rate this lead" survey so it actually helps you, and the source-side changes that shrink the problem. It also covers one thing you should not do — review-gating — because it creates legal risk without solving anything.
The types of junk LSA leads
"Spam" gets used loosely, but the leads that frustrate advertisers fall into a few distinct types, and they are not all treated the same way:
- Robocalls and telemarketers. Automated or sales calls that have nothing to do with hiring you.
- Wrong-number leads. Callers who dialed you by mistake and are not looking for your service at all.
- Out-of-area leads. Real prospects, but outside the geography you actually serve.
- Out-of-service-type leads. People asking for work you do not do — a job type outside your list.
- Price-only tire-kickers. Callers fishing for a quote with no intent to book.
- Repeat or duplicate leads. The same caller generating multiple charged leads.
The important split is between leads that are genuinely not for you (spam, wrong number, out-of-area, out-of-service-type) and leads that are for you but simply did not convert (a tire-kicker in your area for a service you offer). Google treats those two groups very differently when it comes to credit.
What Google's auto-credit will and won't cover
The old process of manually disputing every bad lead ended around July to August 2024. Google now uses a machine-learning auto-credit model: leads are assessed roughly within 72 hours and credits are typically issued within about 30 days, paired with a "Rate this lead" survey that feeds the system. You do not file individual disputes the way you once did.
As a rule of thumb, clear spam, robocalls, and wrong-number leads are often creditable — they are not real service inquiries. What is generally not creditable is a lead where you serve the job type and the area but the customer just did not book, or a job-type or geo mismatch that Google considers your responsibility to have scoped. Note too that some verticals are excluded from the credit system entirely, including healthcare and tax. Because the model, timelines, and eligibility can change, confirm the current specifics against Google Local Services Ads Help.
Set expectations accordingly. Third-party estimates put recoverable spend at only around 6 to 7 percent. That is real, but it is a minority of the waste — which is exactly why cutting junk at the source matters more than chasing credits after the charge.
Use "Rate this lead" accurately and consistently
Because credit is now driven by machine learning, your ratings are training data. If you rate leads accurately and consistently — marking genuine spam as spam, flagging wrong numbers, and honestly scoring bookable leads — the model learns your account and gets better at auto-crediting the right ones. If you rate carelessly or try to game it by flagging everything, you pollute the signal and may weaken the very system that is supposed to protect you. Treat the survey as an ongoing investment, not a chore.
Reduce junk at the source
Source-side reduction is where the biggest gains are, because a junk lead you never receive costs nothing and needs no credit. The levers:
Tighten service categories and service area
Trim your enabled services to the work you actually do and want, and narrow your service area — down to the zip-code level where the tools allow. This directly cuts out-of-area and out-of-service-type leads, which are among the most common junk types.
Refine hours and schedule
Align your hours and ad schedule with when you can genuinely take and work leads. Receiving leads at times you cannot serve them turns real demand into junk on your end.
Sharpen your Google Business Profile
A clear, accurate profile helps prospects self-select. When your categories, description, and service details make it obvious what you do and where, fewer mismatched callers reach you in the first place.
Screen and route calls, block repeat spam
Screening and routing catch junk before it eats your team's time, and blocking known repeat spam numbers where possible stops the same bad caller from charging you again. Fast triage on the rest limits the damage from what does get through.
Junk lead type, creditability, and source-side fix
| Junk lead type | Often creditable? | Source-side fix |
|---|---|---|
| Robocall / telemarketer | Often yes (not a real inquiry) | Screen and route calls; block repeat numbers |
| Wrong number | Often yes | Sharpen GBP so intent is clearer |
| Out-of-area | Depends — geo mismatch often not credited | Narrow service area to the zip level |
| Out-of-service-type | Depends — job-type mismatch often not credited | Tighten enabled service categories |
| Price-only tire-kicker | Generally no (you serve them, no booking) | Set expectations in GBP; fast triage and follow-up |
| Repeat / duplicate | Sometimes | Block repeat spam numbers where possible |
Creditability is not guaranteed for any single lead — the model decides — so treat the middle column as tendencies, not promises, and verify current rules with Google.
What not to do: review-gating
When junk leads pile up, some advertisers try to protect their rating by asking only happy customers for reviews. Do not. Review-gating is risky under the FTC's fake-review rule, 16 CFR 465, effective October 2024, and it does nothing to reduce junk leads anyway. Compliant design asks all customers for reviews. The way to a strong rating is genuine review velocity and responsiveness, not filtering who you ask.
The takeaway: too many spam LSA leads is mostly a source-side problem. Google's auto-credit recovers only an estimated 6 to 7 percent of spend, so the real leverage is upstream — tighten categories and service area to the zip level, refine hours, sharpen your profile, screen and block repeat spam, and rate every lead accurately so the machine-learning credit system works in your favor. Skip review-gating entirely; it adds legal risk without cutting a single junk lead. Reduce at the source, triage the rest fast, and let credit recovery be the backstop rather than the strategy.
Frequently asked questions
Will Google credit spam LSA leads?
Google now uses a machine-learning auto-credit model that assesses leads roughly within 72 hours and issues credits typically within about 30 days. Clear spam, robocalls, and wrong-number leads are often creditable. Leads where you serve the job type and area but the customer simply did not book are generally not creditable. Third-party estimates put recoverable spend around 6 to 7 percent.
How do I stop getting so many junk LSA leads?
Reduce junk at the source: tighten your service categories and service area down to the zip level, refine your hours and schedule, sharpen your Google Business Profile so intent is clearer, screen and route calls, and block repeat spam numbers where possible. Source-side reduction plus fast triage matters more than chasing credits.
Should I only ask happy customers for reviews to offset junk leads?
No. Review-gating — soliciting reviews only from happy customers — is risky under the FTC fake-review rule, 16 CFR 465, effective October 2024. Compliant practice asks all customers for reviews. Review-gating is not a fix for junk leads and can create legal exposure.