Google Local Services Ads (LSA) rarely fails loudly. It fails quietly — a few percent of budget wasted here, a slightly worse booking rate there — and by the time an owner notices, months of margin are gone. Below are seven mistakes we see most often among home-service businesses, and the concrete fix for each. None require spending more; several will lower what you spend per booked job.
1. Judging LSA by cost-per-lead instead of cost-per-booked-job
The dashboard shows cost-per-lead — often cited around $53 on average, ranging roughly $12–$180 by trade and metro. But since a large share of raw leads (third-party estimates suggest ~45%) are unbookable, that number tells you almost nothing about profit. Fix: track spend divided by booked jobs, and compare it to the revenue those jobs produce. That is the number that says whether LSA is working.
2. Leaving service types too broad
LSA spans 70+ home-service categories, and it is tempting to list every service you could conceivably do. The result is a stream of mismatched leads for work you do not want — and many job-type mismatches are not creditable, so you eat the cost. Fix: trim your service types to the work you actually want to run. Your bookable rate rises immediately, with no budget change.
3. Ignoring geography
Treating your whole service area as equally valuable is expensive. Some zip codes produce leads that book; others produce distance-heavy, low-margin, or no-show leads. Fix: review lead-to-booking outcomes by zip and de-emphasize or exclude the weak ones. Zip-level management is one of the highest-leverage moves in LSA.
4. Slow response to leads you already paid for
Because you pay per lead, a call sent to voicemail is money already spent with little chance of return. Responsiveness is also a recognized LSA performance factor, so slow replies hurt ranking too. With the standalone LSA mobile app retired in January 2025, there is no app nudging you on the go. Fix: add an instant acknowledgment (even automated) and an after-hours path so no paid lead goes cold.
5. Not recovering credits
Manual disputes ended in 2024; Google now auto-assesses invalid leads (typically within ~72 hours, credited within ~30 days) using a "Rate this lead" survey for calibration. Owners who never flag leave the estimated 6–7% of recoverable spend on the table. Fix: flag invalid leads accurately and weekly. Remember the limits — job-type/geo mismatches often are not creditable, and healthcare and tax verticals are excluded.
6. Set-and-forget budgets
A static budget starves you in peak demand and overspends in the lull, and pushing budget past the "sweet spot" buys progressively worse leads while cost-per-booked-job climbs. Fix: pace budget to seasonal demand, watch for mid-period burnout, and judge every change by booked-revenue rather than raw lead count. When quality dips, protect spend instead of chasing volume.
7. Risky review practices
Reviews influence LSA performance, so it is tempting to ask only your happy customers. That is review-gating, and it runs into the FTC's fake-review rule (16 CFR 465, effective October 2024). Fix: ask all customers for reviews, consistently and promptly. Compliant review-building is also better long-term practice, and since around July 2025 LSA reviews are managed through your Google Business Profile, so keep that linkage clean.
Quick self-audit
| Mistake | Symptom | Fix |
|---|---|---|
| Wrong metric | Judging by cost-per-lead | Track cost-per-booked-job |
| Broad service types | Mismatched job leads | Trim to wanted work |
| No geo management | Budget in weak zips | Exclude poor performers |
| Slow response | Voicemails, no callbacks | Instant + after-hours reply |
| No credit recovery | ~6–7% left unclaimed | Flag invalid weekly |
| Static budget | Burnout or overspend | Pace to demand + sweet spot |
| Review-gating | FTC exposure | Ask all customers |
Each of these is small on its own. Together they are the difference between an LSA account that quietly leaks margin and one that turns paid leads into profitable booked work. Run the self-audit, fix the two or three that apply most to you, and re-check your cost-per-booked-job in a month.
Frequently asked questions
What is the most common LSA mistake that drains budget?
Judging LSA by cost per lead instead of cost per booked job. Because a large share of raw leads, estimated near 45 percent, are unbookable, cost per lead tells you little about profit. Track spend divided by booked jobs and compare it to the revenue those jobs produce.
How can I stop paying for the wrong LSA leads?
Tighten the inputs. Trim service types to the work you actually want, manage geography at the zip level so weak areas do not absorb budget, and keep your profile accurate. Many job-type and geo mismatches are not creditable, so preventing them beats disputing them.
Is it safe to ask only happy customers for reviews to boost LSA?
No. That is review-gating and runs into the FTC's fake-review rule (16 CFR 465, effective October 2024). Ask all customers consistently and promptly. Since around July 2025 LSA reviews are managed through your Google Business Profile, so keep that linkage clean.