What's the right way to compare Local Services Ads against paid search?

The two channels do not count the same event

A Local Services charge corresponds to an actual contact: a call, a message, a booking. A search conversion may be a form fill, a call of qualifying duration, or a click on a phone link. Those are not the same thing, and neither is a job.

Comparing cost per lead across them is therefore comparing two different units. The only common unit both channels can be carried to is a completed job with revenue attached, which lives in your field service system, not in either ad account.

Build the comparison on one table

The structure is simple even when the plumbing is not. One row per charged lead or attributed click, with the source, the date, the cost side, and the operational outcome joined on.

• Source and subtype. LSA call, LSA message, search call, search form. Blending them hides the differences you are looking for.
• Contact outcome. Answered, qualified, booked, completed, cancelled.
• Revenue. Completed invoice value, not estimate value.
• Lag. Days from lead to completion, so you know how long to wait before judging a period.

Respect the lag or you will misread everything

Emergency work completes in days. Replacement and project work completes in weeks. If you compare a channel that skews emergency against one that skews project on a thirty-day window, the project channel will look worse than it is, every single time.

The fix is a cohort view: group leads by the week they arrived and let revenue accrue against that cohort over time. It is more work to build and it is the only version that does not lie. This is the core of what marketing intelligence has to get right.

Then ask the harder question

Even a correct comparison does not tell you to move budget. A channel with lower revenue per unit of spend may still be incremental, may reach customers the other cannot, or may be capacity-limited rather than efficiency-limited.

Read the comparison as one input into a capacity and margin decision, alongside how much work your crews can absorb and what a job in each channel costs to deliver. The best-performing channel is not worth more budget if you cannot staff the jobs it produces.

It also matters what each channel does to your customer base over time. A source that produces one-off emergency work looks different at three years than a source that produces maintenance customers, and neither the ad account nor a single-quarter comparison will show you that.