Revenue per lead by channel: home services attribution that holds up

Why "the phone rang" is not an attribution model

Home services live and die on the phone call. Across the U.S., roughly 25% of inbound home-service calls go unanswered, and a missed call is a lead with a source and no owner. Most contractor CRMs still bucket every inbound as "phone call" with no channel attached, which means every channel that generates a phone lead gets credit for exactly nothing.

The pattern is consistent in the accounts we audit: the channels that show the highest cost-per-lead in a dashboard are often the ones doing the real work, because their leads convert at a higher rate than the cheap form-fill channels the dashboard rewards. Attribution that stops at "lead volume" punishes the channels that close, and it is the single biggest reason home-service owners keep funding the wrong media.

This is not a software problem. It is a plumbing problem. The capture layer has to be built before the reporting layer can mean anything.

Revenue per lead by channel: the honest benchmark

No two trades have the same ticket size, so revenue-per-lead benchmarks only hold within a trade. A plumbing service call might close at $400 while a roof replacement closes at $11,000, and comparing their cost-per-lead directly is meaningless. What holds up across trades is the shape of the data: which channels cost the most per lead, and which convert the highest share of what they send.

Sources: Searchlight Digital 2026 benchmark (888 contractors, 126,650 leads); Google LSA support documentation; PipelineOn home-service benchmarks 2026. CPL ranges are averages across U.S. metros and move with trade and ZIP code.

The number that matters most is not in this table. It is the revenue-per-lead figure, and that requires you to attach a job value to each closed lead and divide by the channel. A channel sending 10 leads at $120 each that close at 40% produces more net revenue than a channel sending 30 leads at $35 each that close at 8%, and a last-click dashboard will fund the second one all day.

Where attribution breaks for every channel

1. Google LSAs: the lead is reported, the revenue is not

LSA is the easiest channel to attribute because Google invoices per valid lead and even allows disputes for invalid ones. But the revenue side still goes dark: an LSA lead that calls your shop directly and books does not automatically close the loop back to the LSA record unless your team logs it. Google reports a lead; only you can report a job.

2. Search Ads: the click is counted, the call is not

Google Ads counts clicks and, if you enable call reporting, calls from the ad. What it routinely under-counts is the phone number on your landing page: a buyer who clicks, reads, then dials the number on the page is often recorded as a click with no call event, so the conversion never fires. That under-count is typically 20% to 40% of phone conversions, which is why every landing page needs a call-tracking number tied to the campaign, not your main line.

3. Meta: the view-through problem

Meta's default reporting leans on view-through conversions, where an impression with no click is credited when a conversion happens later. For a home-service business, that produces two opposite errors. It over-credits Meta when a searcher saw your ad once and then Googled you, and it under-credits nothing, because the phone call Meta did influence is invisible. The fix is to read Meta as an awareness channel for home services, and to weight it by branded-search lift rather than by in-platform conversions.

4. Organic and Local 3-pack: the highest-converting channel you never tagged

Organic search and the Local 3-pack consistently produce the highest close rate in home services, because a buyer who finds you on the map already trusts you enough to call. But almost no contractor tags these calls. The result is the classic attribution inversion: the channel that closes best looks like it produces nothing, so budget shifts to paid, and the whole flywheel slows down.

5. Direct, referral, and repeat: the base load you are already standing on

Repeat and referral revenue is the most profitable revenue in the business, and it is invisible in every channel dashboard because there is no media spend attached. It is still revenue per lead. If you only measure paid channels, you will under-invest in the follow-up, review, and referral systems that make the base load grow, because those systems show up in no paid report.

The three fixes that make attribution hold up

Fix 1: unique tracking numbers per channel, routed to one line

Assign a distinct call-tracking number to each channel and surface it dynamically: one number on the LSA listing, one on your Search Ads landing page, one on your Google Business Profile, one in your email signature. When a buyer dials, the platform routes the call to your main line and records the source. This single change is what surfaces the 25% to 30% of revenue that phone leads represent.

Fix 2: UTM parameters that survive the form

Every paid link, every email, and every listing should carry UTM parameters, and your booking form should pass those parameters through as hidden fields. A form that does not carry source means every form lead reverts to "unknown." The technical lift is small; the discipline is what most shops lose by month three.

Fix 3: score first-touch, report last-touch, and read both

First-touch attribution tells you what started the relationship. Last-touch tells you what finished it. For home services, the honest read is both: first-touch shows which channels fill the top of the funnel, and last-touch shows which ones actually close. When the two diverge sharply for a channel, that is the signal to investigate, not the signal to cut.

The number that actually matters

Stop reporting cost-per-lead as the headline metric. Report revenue per lead, and hold every channel to it. A channel is worth funding when the revenue it produces, minus its media cost, beats the next-best channel you could fund instead, at the same total spend. That is the only comparison that survives contact with a real profit and loss statement.

And remember the speed variable: the median home-service business responds to a lead in 42 minutes, while the top 10% call back in under 5 minutes and close at roughly 21 times the rate. Attribution is moot if the lead goes cold before anyone picks up the phone. A well-attributed channel with a slow response is still a losing channel.

Related reading

• What 10 extra minutes of lead response time costs a home service business: the revenue math on response speed, which is the other half of the attribution equation. A lead you can attribute but never answer earns nothing.
• Why your Google Local Service Ads aren't spending (and the 6 things to check first): if the LSA channel in your attribution report shows zero, start here before you trust the rest of the dashboard.