What Paid Media Channels Still Deliver Quality Leads? The Surprising Answer.

---

In Home Services – Your Cost Per Lead Is Up 4x Since 2020. The Fix Isn’t a New Paid Media Channel.

$23.7 million in home services paid media ad spend across seven years actually shows about channel mix, Local Services Ads, and where the advertising budget is really lost.

If you’ve looked at your ad spend over the last six years and felt like you’re paying four times more for the exact same phone calls, you’re not crazy. The math proves it, and the cost for quality leads through paid media has skyrocketed since 2020.

Here is what our own book looks like. This is every dollar we’ve managed for home services clients only, from January 2020 through September 2026. More than 408,000 tracked conversions, across search, Local Services Ads, Bing, programmatic, and paid social.

Year	Blended cost per lead
2020	$19.11
2021	$29.81
2022	$59.30
2023	$63.20
2024	$67.21
2025	$75.79
2026 (through September 20, 2026)	$85.29

_{(we use blended cost per lead here, but I will explain later why this is a faulty metric)}

Four and a half times higher in six years. And 2026 is the steepest year of all.

If you run a home services company, you have felt this. Your ad spend went up, your lead counts probably didn’t, and somewhere deep down, you started to wonder if your agency was doing their job. I understand the reaction, and believe me, we here at City Ranked have fielded many questions around this paid media trend.

But I want to show you what’s actually happening, because most of the advice being sold to contractors right now is built on a picture of the market that stopped being true about two years ago. And the fix that does work, is not the one most agencies are talking about, or fixing for that matter.

The cheap(er) paid media channel is gone

Walk into any marketing conference for the home services trades and you will hear some version of this: Google Ads is expensive, so shift your money into Local Services Ads. You only pay per lead, you get the verified checkmark, and you sit above everything else on the page. Cheap, high-intent shoppers. The paid-media marketing is handled, right?

That was excellent advice in 2021. But our data* shows it’s no longer true, and we have the spend to back it up.

Year	Google Ads CPL	Local Services Ads CPL
2020	$22.54	$26.27
2022	$84.51	$41.90
2024	$70.78	$62.77
2025	$73.31	$80.96
2026	$79.29	$96.97

The above represents $7.2 million of Local Services Ads spend in home services across our book of business. In 2022, LSAs cost half what search cost. In 2026 they cost 22% more.

Local Services Ads went up 54% in under three years. Nobody is writing about this, because the LSA playbook is still what agencies lead with in pitch meetings. It’s easy to explain and it sounds like a discount because you’re paying for ‘qualified’ leads willing to call you directly.

Here’s why it moved. LSA pricing isn’t fixed, it’s a local auction that adjusts based on how many contractors in your service area are in it, and every year more of them join the auction. Google also factors in who gets shown by responsiveness and review profile, not just budget, so the cost of a lead in a competitive metro drifts upward as the field gets better at the underlaying mechanics of what is being judged. As a result, because you’re paying per lead, rather than per click, you feel every bit of that auction drift immediately, usually as increased conversion costs, or worse positioning resulting in less conversions.

“Your LSA costs went up, or your calls dried up. It looks like your agency let something slip. It didn’t, the auction got harder. And Google isn’t done with making changes.”

The reasonable reaction to this as a business owner is to assume your agency is letting things slip. After all, costs for the same exact calls have increased (a lot), or worse, calls from LSA’s have dropped. Both are detrimental to your business, so I understand the desire to find out why quickly, but Google isn’t done with changes yet.

Starting October 1, 2026, Google changes what counts as a billable lead in Local Services Ads, a change that will push these numbers up again for reasons that have nothing to do with the auction. I’ll come back to it, because for most of you it’s the most consequential thing in this article.

Adding more paid media channels doesn’t fix it either

The next thing you’ll be told is to stack paid media products. Run LSAs, and search, and Performance Max, and retargeting, and paid social, all at once, and your blended cost per lead comes down because the cheap channels average out the expensive ones.

I believed a version of this. When I went looking in our own data to prove it, I was surprised when I got the opposite answer.

Here is every home services account we’ve run from 2024 through September 2026 — $10.3 million across 93 accounts — sorted by how many paid channels were actually running in a given month.

Paid channels running	Account-months	Spend	Blended CPL
1	244	$675,833	$72.53
2	709	$7,628,635	$76.54
3	96	$1,989,626	$97.16

More channels = higher cost per lead. Straight line, wrong direction.

I ran the data a second way to make sure, because comparing different accounts to each other is a weak test; bigger, better-funded companies tend to run more channels, and that alone could skew it. So I looked at the same account in the same year, comparing its single-channel months against its multi-channel months. Multi-channel came out cheaper in 19 of 42 cases, less than 50% of the time.

Let me be pointed about what this means. We sell multi-channel programs. If channel stacking reliably drove cost per lead down, that would be a very convenient thing for me to publish, but the data does not support it.

This article isn’t an argument for reasons to only run one channel. Some real world reasons to be in more than one channel, coverage when a particular channel has a bad week, reaching people who don’t use that channel, not having your entire lead flow reliant on one algorithm. Those are all good reasons. “It’ll lower my cost per lead” is not one of them, and if an agency tells you it is, ask to see the data they have that supports it.

The bigger problem: cost per lead is the wrong number

I’m going to show you the thing that many agency blogs leave out, and as a business owner this matters a lot when it comes to paid media campaigns and conversion tracking.

How leads are your being counted? We count a qualified lead more strictly than most agencies do, and we’ve been doing strict conversion tracking since we started in 2011. For us, at least for most of the data presented in this blog (as we’re now using the AI-powered Skout IVR for lead qualification), a conversion has to be a first-time caller, nobody who has called in the last sixty days from the same channel, and the call has to run two minutes or longer. Somebody dialing a wrong number or hanging up after fifteen seconds isn’t a lead. And we do not count calls that come in through brand campaigns as new-customer conversions. We report brand separately, because it doesn’t belong in the new-customer column.

Why that last rule regarding brand-conversions matters more than it sounds.

Blended Cost-Per-Lead (and why we don’t use it)

A brand campaign is when you bid on your own company name so that when somebody searches for you specifically, a competitor doesn’t sit above you. It’s cheap, it’s worth doing, and the leads are almost never new customers. They’re people who already know who you are; current customers, past customers, referrals, somebody who saw your truck, conversions your agency should not be counting into your new lead totals, or your cost-per-lead.

Here is our current campaign data, agency-wide for the home services vertical:

Campaign Type	Cost per lead	Share of all reported conversions
Brand campaigns	$13.54	34.6%
Everything else (non-branded)	$124.58	65.4%

Look at what this does to the number many business owners look at, what is the Cost-Per-Lead that my agency is driving, a real number that many business owners judge their agency on.

• Blended cost per lead, the way the industry reports it: $69.24
• Cost per lead for an actual new customer: $98.73

The blended cost per lead understates what a new customer really costs by 43%.

Think about what that makes possible. An agency can move budget toward brand campaigns, the cheapest leads on almost any campaign, watch the blended number drop, and put that on your monthly report as an improvement. Blended CPL down 20%, that number is real, but it’s the wrong number to measure. Your cost per lead genuinely fell, but you did not acquire one additional customer, because you just bought more conversations with people who were already going to call you regardless. I’m telling you this because this standard metric is easy to move without moving your business, and most people reporting it are looking at surface level metrics.

There’s a second thing in that table worth looking at. Non-branded leads, the ones from strangers, the ones that actually grow your revenue cost $124.58. In one multi-state pest control platform we run, non-branded leads are at $146.72 while their brand leads sit at $19.73. That’s more than seven times the spread inside the same account, in the same week.

Ask your agency for non-branded cost per lead, separated out, as this will give you a clearer picture of what you spend to receive one qualified lead to your business.

The leads you already paid for and never get a chance to sell

So if you can’t switch to a cheaper paid media channel, and adding channels doesn’t help, and the metric most agencies report is misleading, what’s left for a business to do?

Local Services Ads report something most channels don’t: how many of your leads called, and how many of those calls actually connected to a human being at your business.

Across 38 home services locations run by 10 companies, our clients spent $200,250 and received 2,209 charged leads from 3,169 phone calls.

• 447 of those calls, 14.1%, were never connected to a live sales or customer service representative.
• One in seven. Paid for, delivered to the business, rang, and nobody picked up.

When I first tackled this blog post, I looking for data that pointed to something bigger, but didn’t find it. My theory was that answer rates factor big into cost per lead, that the businesses who pick up the phone rank better in Google’s LSA auction, get shown more, and therefore pay less per lead. Sorted generally, our data looked like it confirmed that: the locations answering 93% of calls got far more leads per dollar than the ones answering 79%.

I wanted to make sure, so I controlled for location size, and most of the effect I saw initially disappeared. Larger budget locations pay more per lead regardless of how they answer the phone, and they were clustered in the middle of the answer-rate range. Among the larger ones the relationship was essentially flat when comparing locations between a high 70% to low 90% answer rate.

Among our smaller businesses, though, there was something real. The ones answering around 93% of calls got about 12% more leads per dollar than the ones answering around 80%. Our average location across all 38 answers 85.9% of calls.

So what’s the ceiling? Not 92.6%, which is just where our best quarter happen to sit. We have locations answering 100%, including one running 96.6% across 88 calls in a single period, this is not a small-sampling, it’s an operational standard some businesses already have in place, every single phone call is answered, no exceptions.

Closing from our average to our best quarter is worth somewhere around 6 to 8% more leads on the same budget. Closing all the way to 100% of billable LSA calls is worth more like 13 to 16% more leads. All for the same budget.

Our clients, on average, answer 86% of their billable LSA calls. Closing that to 100% projects to approximately 16% more leads on the same ad budget.

That is not the one-third improvement the raw sort suggested, and I’m not going to claim it is. But up to 16% more customers for zero additional ad spend is not nothing, and it’s the only lever in this article that doesn’t cost money to pull.

Our data shows that one in seven calls your LSA advertising generated never reached a person. You bought those. They were as real as the ones that got answered. They called a competitor instead.

And on October 1, Google starts billing you for them

Up to now, a missed Local Services Ads call has been an ambiguous loss. Sometimes you were charged, sometimes you weren’t, and most operators never looked closely enough to know which.

That ends this month. Google has notified advertisers that as of October 1, 2026, it is changing how and when you are charged for call leads from Local Services Ads. Two changes matter:

• A missed call during your business hours becomes a charged lead if the caller stays on the line longer than 20 seconds.
• Follow-up calls between you and that customer within 15 days count as a single charge, rather than billing separately each time.

This is what makes “100% of billable calls” a target you can actually hold rather than a slogan. Google has now drawn the line for you: pick up within twenty seconds during the hours you publish, and there is no unanswered billable call on your invoice. And something for you home service business owners with an IVR (Interactive Voice Response – Phone Tree), if you’re IVR is longer than 20 seconds, and the person hangs up because they waited for a representative too long, you’ll most likely still get billed.

Over the years Google has rewarded being consistent as a business. They have spent years weighting Local Services Ads toward businesses that respond, responsiveness already affects who gets shown. This closes the loop on the money side. If you tell Google you’re open and a homeowner calls and waits, Google has done its job. Whether that call turned into a conversation is now your problem, and they have now priced that accordingly.

If you run Local Services Ads, there are two things worth doing before the end of this month. Check that the business hours in your LSA profile match the hours somebody is actually there to answer, if you’ve got hours set that nobody in your office covers, you are about to start paying for that. And find out what your real answer rate is. Most operators have never seen the number.

That is the cheapest inventory left in home services marketing, and it isn’t in an ad account. It’s in your phone system, your staffing at 6pm on a Friday, and whether anything at all happens to a lead that comes in after hours.

---

What this means for how you buy marketing

If you take one thing from all of this, take this: the ad account is no longer where the leverage is.

Media buying still matters. Somebody has to keep non-branded campaigns pointed at the work you actually want; replacements, installs, and recurring service, not the cheap stuff that fills the schedule and starves the P&L. Somebody has to separate brand from non-brand so you know what you’re really paying. That should be standard now, not a competitive edge.

“The competitive edge isn’t the ad anymore. It’s whether the phone gets answered at 7pm on a Saturday, or 10am on a Monday, and whether whoever picks up books the job or takes a message.”

The edge is in what happens to the lead after you’ve bought it. Whether the phone gets answered at 7pm on a Saturday or even 10am on a Monday. Whether the person who answers books the job or takes a message. How fast the follow-up goes out to the ones who didn’t book. Whether anybody ever finds out which of those leads turned into revenue, so next month’s budget goes where the money actually came from.

Those are the post-click channels with room to influence buying decisions. City Ranked runs about ten of them in the systems we build and manage, and every one of them lives downstream of the paid media click.

The October billing change is worth reading as a signal, not just a cost. Google is moving toward paying for readiness, putting the businesses that answer in front of homeowners, and charging the service regardless if you engage or not. That direction isn’t going to reverse. The platforms have more data on how you handle a lead than they’ve ever had, and they are increasingly willing to price it as such.

Here’s the part most agencies can’t say, because they do not have the technical infrastructure to handle it, and none of these downstream channels can run until the foundation is in.

• A connected CRM so leads can be tracked to booked revenue.
• Call tracking that correctly attributes to the right channel.
• Text messaging properly registered so your follow-ups don’t get filtered.
• Analytics that track each stage of your sales pipeline.

We won’t start the levers until the above is done, and we won’t take responsibility for a lead goal without it, not because we’re sticklers about process, but because a lead-handling system built on data points you can’t trust will produce confident reports about nothing.

That’s usually the first real conversation we have with an business owner. Not which channel, not what budget. Whether the connections under the marketing is good enough to measure what we’re about to do.

If your cost per lead has been climbing and you’ve been blaming the ads, it’s worth finding out whether that’s actually where it’s going.

---

How we measured this

All figures come from City Ranked’s own client accounts, January 2020 through September 19, 2026, $23.7 million in managed ad spend, and more than 408,000 tracked conversions across Google Ads, Google Local Services Ads, Microsoft Ads, programmatic display and paid social.

Conversion definition. A conversion is a first-time caller on a 60-day lookback whose call lasted two minutes or longer. Brand-campaign conversions are tracked and reported but are excluded from new-customer counts. This is a stricter definition than most published benchmarks use, many count form fills, short calls and repeat callers, so our figures are not directly comparable to industry averages and will generally read higher than studies using looser criteria.

Channel-count analysis covers home services accounts only, 2024 through September 2026: 93 accounts, $10.3 million, grouped by the number of paid channels with spend in a given account-month.

Answer-rate analysis uses Google Local Services Ads reporting across 38 active home services locations operated by 10 businesses in a single reporting period, covering $200,250 in spend, 3,169 inbound calls and 2,209 charged leads. Answer rate is connected calls divided by total calls as reported by Google. The unconnected-call figure (447 calls, 14.1%) is a direct count.

Local Services Ads billing change. Google notified advertisers in August 2026 that effective October 1, 2026 it is updating how and when call leads are charged: missed calls during business hours become charged leads when the caller remains on the line longer than 20 seconds, and subsequent calls with the same customer within 15 days are billed as a single lead. The $40,500 figure cited above is an upper bound calculated as unconnected calls times our average charged-lead cost, and assumes every unconnected call would qualify under the new rule; the actual figure will be lower. Source: Google advertiser notice, reported by Search Engine Roundtable, August 25, 2026.

---