Why "Just Cut The Budget" Is the Wrong First Move
The instinct when a CFO or owner sees rising ad costs is to cut the budget across the board. This is close to the worst way to solve the problem, because an even, proportional cut reduces spend in your best-performing ZIP clusters exactly as much as your worst-performing ones. The right sequence is: find the waste, verify it's actually waste, then cut only that.
Step 1: Separate Three Distinct Types of Waste
Geographic Mistargeting
This is the Location Sciences finding in practice: campaigns configured with radius or DMA-level targeting that don't match the business's actual serviceable area.
Fix: Audit every active campaign's geographic settings against your actual dispatch radius. This alone often recovers 10–15% of spend with zero effect on qualified lead volume.
Invalid Traffic (IVT)
Lunio's 2024 analysis put invalid traffic at 8.5% of all paid traffic, driven by bots, click farms, and competitor click fraud.
Fix: Enable platform-native invalid click protection, layer in third-party click fraud detection, and review IP-level click patterns for repeat non-converting clicks.
ZIP-Cluster Underperformance
The waste that's hardest to see: budget flowing into ZIP clusters that, under a proper geo-holdout, show flat or negative incremental lift. This is the self-attribution inflation problem detailed in The Hidden Cost of Platform Self-Attribution.
Fix: Run geo-holdout tests by ZIP cluster (see The Complete Guide to Home Services Marketing Attribution) and reallocate budget away from clusters that fail.
| Bucket | What to check | Typical recoverable % |
|---|---|---|
| Geographic mistargeting | Campaign geo-settings vs. dispatch radius | 10–15% |
| Invalid traffic | Click timing, IP repetition, IVT flags | 5–10% |
| ZIP-cluster underperformance | Geo-holdout test per cluster | 10–15% |
Step 2: Fix Geographic Targeting First
Before touching anything statistical, fix the geographic settings. This requires no testing period and no risk to lead volume. For franchise systems, this also solves the co-op ad fund complaint about leads generated outside a franchisee's territory — see the Franchise Co-Op Ad Fund Optimization Playbook.
Step 3: Layer in Invalid Traffic Protection
For home services accounts with meaningful spend (generally $5,000+/month per channel), a dedicated third-party click fraud tool typically pays for itself within the first month.
Step 4: Run Geo-Holdout Tests Before Cutting Any ZIP Cluster
This is the step most operators skip. The only way to know for certain whether a low-ROAS ZIP cluster is truly wasted spend is to hold the channel out in that cluster for 2–4 weeks and compare booked-job volume against a matched control. Only cut spend in clusters that fail this test.
Worked Example: Finding $2,000/Month in a Single ZIP Cluster
Consider a hypothetical HVAC company spending $15,000/month across Google Ads, Meta, and Local Service Ads. A geo-holdout audit run cluster by cluster might reveal that one lower-density, higher-competition suburban cluster is absorbing roughly $2,000/month on paid search, yet a holdout test shows almost no incremental booking lift there — bookings occur at nearly the same rate whether the campaign runs or not, because organic search and repeat business already dominate demand. Reallocating that $2,000/month toward a cluster where the same methodology shows strong incremental lift doesn't just stop the waste, it actively improves total bookings for the same total spend.
The Five Most Common Sources of Wasted Spend, Ranked by Frequency
- Invalid and bot traffic on display and social campaigns — Lunio measured this at 8.5% of all paid traffic, contributing to over $71 billion in global wasted spend in 2024.
- Overlapping geo-targeting across channels in the same ZIP cluster — bidding against your own incremental demand ceiling.
- Broad-match keyword drift on paid search, sending spend to searches with no commercial intent.
- Dayparting misalignment with call-center or dispatch capacity — leads that go unanswered still cost the same as leads that convert.
- Low-intent retargeting audiences run past their useful window, past the point where the customer has already converted elsewhere or lost interest.
What to Do With the Recovered Budget
Reinvest the recovered budget into the ZIP clusters and channels that passed the holdout test, and re-run the audit quarterly, since seasonal demand shifts change which clusters are truly incremental over the year.
FAQ
Is 30% a realistic target for every home services business?
It's a realistic aggregate across the three waste buckets, but the actual mix varies by business.
Will cutting wasted spend reduce my lead volume?
Not if the waste audit is done correctly, since the spend being cut wasn't producing incremental bookings in the first place.
How long does a geo-holdout test need to run?
Typically 2-4 weeks, depending on sales cycle length and lead volume in the tested cluster.
What's the difference between invalid traffic and low-quality leads?
Invalid traffic never had any chance of converting. Low-quality leads are real people who clicked but weren't a good fit.
Should franchise co-op ad funds run this same audit?
Yes, and arguably more urgently, since co-op funds spend across many franchisee territories simultaneously.
Do I need to pause campaigns to run a geo-holdout test?
You pause the channel only in specific test ZIP clusters, not account-wide, while running normally in matched control clusters.
How is this different from an audit an ad agency would run?
Most agency audits rely on platform-reported metrics. This framework adds the incrementality layer platform metrics can't provide.
Try the Wasted Ad Spend Estimator
See a directional waste-dollar range for your budget using the same Lunio and Location Sciences benchmarks cited above.
Get your free ZIP report