Why shared lists cap out (and why more of them will not fix it)
A shared lead is a homeowner's phone number sold to three or four contractors at once. The marketplace makes money on the sale, not on whether you book the job, so its incentive is to sell the same name as many times as it can. That is the whole model. You are not buying a customer. You are buying a coin-flip against your competitors, and you are paying for the entry fee whether you win it or not.
Two things break because of it. First, price. When four crews call the same roof in the same hour, the homeowner learns fast that estimates are negotiable, and the cheapest number tends to win. Your margin is the casualty, and you cannot out-work a race you were entered into against your will. Second, quality. Shared platforms reward volume, so tire-kickers, price-shoppers, dead numbers, and people who filled out a form by accident get sold right alongside real jobs, and you pay for all of them at the same rate.
Buying more shared leads does not escape this. It doubles down on a channel where you own nothing. The day you stop paying, the flow stops cold, and you have built no asset: no ranking, no reviews under your name, no list of past customers you can call again. You have rented attention and have nothing to show for the rent. That is the difference between an expense and an investment, and shared lists are pure expense.
There is also a ceiling problem. Even if you tolerate the price fights, a marketplace only sends you the leads it decides to send you, in the areas and volumes it chooses, at the prices it sets. You are a tenant in someone else's storefront. Scaling means paying more per lead as competition on the platform climbs, not building something that gets cheaper as it grows.
The contractors who get off this treadmill do not find a secret better marketplace. They move spend into channels where the inquiry arrives addressed to them alone. The rest of this guide walks those channels, what each one realistically feeds, and how the economics actually compare on the number that matters.
The channels that feed exclusive contractor leads
Every source that sends you leads falls into one of two buckets: channels you own, and channels you rent. Owned channels keep working after you stop paying and send the inquiry to you alone. Rented channels stop the day the invoice stops. Here is the honest lay of the land.
| Channel | Exclusive? | Own or rent | What it feeds |
|---|---|---|---|
| Your ranked website | Yes | Own | Steady inquiries for the jobs you want, built to compound |
| Google Business Profile / map pack | Yes | Own | Near-me searches with high buying intent |
| Local Services Ads (LSA) | Yes | Rent | Screened, pay-per-lead calls at the top of results |
| AI search answers | Yes | Own | Referrals from ChatGPT, Perplexity, and AI Overviews |
| Referrals and past customers | Yes | Own | Your highest-closing, lowest-cost leads |
| Shared marketplaces (Angi, etc.) | No | Rent | Resold names, priced down, high tire-kicker rate |
Notice the pattern. Everything exclusive is either owned outright or, in the case of LSA, at least sold to you alone and charged per real lead you can dispute. The one non-exclusive row is the one you are trying to leave. The move is not to find a better version of that row. It is to shift budget up the table over time, out of resold names and into flow that is yours.
You do not have to fire the marketplace on day one, and you probably should not. Most contractors keep a shared or LSA feed running to cover the calendar while the owned channels warm up, then throttle the shared spend down as the map pack and site start producing their own inquiries. That transition, from renting resold names to owning your flow, is the whole game. The channels below are the pieces of it.
Owned channels: your site, the map pack, and AI answers
The three channels that build an asset are your website, your Google Business Profile, and your visibility inside AI-search answers. This guide frames the lead outcome, not the mechanics: the how-to of ranking a site, optimizing a Google Business Profile, and getting cited by ChatGPT each have their own playbook. Here we stay on what each one feeds, what it costs you, and why it is worth the wait.
A ranked website is the only channel that sends you a steady stream of inquiries for the exact jobs you want, day and night, without a per-lead charge. Competitive terms typically take 4 to 9 months to move, and a real contractor build usually needs a cluster of pages (94 or more is typical for a full trade site) so you show up for every service and city you cover, not just your homepage. A homeowner searching your specific trade and service is a warmer lead than a marketplace name, because they picked you off the page instead of being sold to you in a batch. It is slow to start and hard to take away once it is running, which is exactly the profile of an asset.
The map pack (the top-3 local results with the map) catches the highest-intent searches there are: someone typing your trade plus "near me" is ready to call, often that day. Reviews, complete categories, and an accurate profile decide whether you land in that top 3. It costs nothing per lead, it feeds the phone with people who already chose to look for you, and it compounds with every job you close, because every job is a chance at another review that lifts you higher.
AI-search visibility is the channel most agencies still ignore, and it is the one we lead with. When a homeowner asks ChatGPT or Perplexity for a contractor, or Google prints an AI Overview above the normal results, being the name that gets cited sends you exclusive referrals your competitors are not even chasing yet. There is no per-lead auction and no shared name. It is early, the field is thin, and that is precisely why it is worth claiming now instead of waiting until every roofer in your market has caught on.
The through-line on all three: you build them once, they keep sending leads addressed to you alone, and their cost per lead falls toward zero as they mature. That is the opposite of a shared list, where the last lead costs exactly what the first one did.
Rented-but-exclusive: Local Services Ads and pay-per-lead
Local Services Ads sit at the very top of Google, above the regular ads, wearing the green Google Guaranteed badge. They are rented (you pay per lead) but exclusive: a lead that comes through LSA is sold to you, not auctioned to four crews at once. That single difference makes them a different animal from a shared marketplace, even though both charge per lead. On LSA you are paying for a call that is yours. On a shared list you are paying for a chance at a call that three other crews are also paying for.
Two things make LSA worth a line in the budget while your owned channels warm up. First, you can dispute and get credited for junk leads: wrong service, spam, someone clearly outside your area, a robocall. You are not eating the cost of every bad call the way you do on a shared list, where a dead number is just money gone. Second, the green badge itself carries trust with homeowners, and it sits above everything else on the page, so the leads that come through it tend to close at a higher rate than a cold marketplace name ever will.
LSA is not free of downside, and it would be dishonest to pretend otherwise. You pay per lead, the price per lead moves with your market and your trade, and it stops the moment you stop funding it, same as any rented channel. The account setup, the screening requirements, and the bidding and budget mechanics are their own topic with their own playbook. For the purpose of this guide, the point is simpler: treat LSA as the fastest exclusive channel you can turn on. It produces this month, while your site and profile take their 4-to-9-month climb.
The right way to use it is as a bridge, not a destination. Fund LSA to keep the calendar full now, watch your true cost per booked job (not just the cost per lead), and taper the spend as the owned channels start carrying more of the load. Rented flow that turns out cheaper than a shared list, and exclusive on top of that, is a fair trade for the runway it buys you while the assets compound.
The referral loop most contractors leave on the table
Your past customers are the highest-closing, lowest-cost leads you will ever get, and most contractors do nothing systematic to work them. A referred homeowner already trusts you before the phone rings, because someone they know already vouched for the work. They price-shop less, argue over the estimate less, and close faster than any paid lead. This is exclusive flow that costs you almost nothing per lead, and it is sitting in your customer list right now, unworked.
The reason it stays on the table is that referrals get treated as luck instead of as a system. Luck is not repeatable, and you cannot budget around it. A loop is repeatable, and you can. The loop has three moving parts: ask at the right moment, make the ask effortless, and stay in front of past customers so you are the name they hand over when a neighbor asks who did the roof.
- Ask at the peak. The best moment for a referral or a review is the day the job passes final inspection and the customer is happiest, standing in front of finished work. Three weeks later the feeling has cooled and the ask feels like an imposition. Wire the request into your closeout, not into a follow-up you will forget.
- Make it one tap. Text a direct review link so leaving a review takes ten seconds, not a hunt through Google. Every review you bank feeds the map pack, which feeds more exclusive leads, so your referral habit and your rankings compound each other instead of running as two separate chores.
- Stay in the pipe. A short seasonal note (gutter season before the leaves fall, a pre-summer AC check, a roof look-over after a storm) keeps you top of mind so the referral goes to you instead of the last truck the neighbor happened to see.
None of this requires a marketplace or a per-lead fee. It requires a habit, wired into how every single job closes out, run the same way every time. Contractors who install this loop find that referrals and reviews quietly become a meaningful share of their booked work, and it is the share with the fattest margin, because you paid nothing to generate it and the customer showed up ready to sign.
How the economics actually compare
The honest comparison is not cost per lead. It is cost per booked job, because a cheap lead that never closes is expensive, and a pricier exclusive lead that books is a bargain. Shared lists look cheap on the sticker and turn expensive the moment you factor in all the leads you paid for that three other crews also called, the tire-kickers you paid for, and the dead numbers you paid for. The sticker price is the smallest part of the real cost.
Run the math on your own numbers with a simple frame:
- Cost per lead: what you pay to generate one inquiry from a channel.
- Close rate: the share of those inquiries that turn into signed jobs. Shared leads close low because you are fighting on price against crews holding the same name; referrals and exclusive leads close high because nobody else is on the call.
- Cost per booked job: cost per lead divided by close rate. This is the number that actually leaves your account, and it is the only one worth comparing across channels.
- Job value: price the whole thing against what a job is truly worth to you over its lifetime, including the referrals and reviews a happy customer sends back, not against the lead price alone.
When you run it that way, shared lists usually lose. A low sticker price times a low close rate times a lead that got resold three times is a bad deal dressed up as a cheap one. Exclusive channels win on the number that pays your bills, even when their cost per lead looks higher on paper. The cost ranges vary a lot by trade and market, and a fuller breakdown of what contractor leads actually cost in 2026, channel by channel, is worth studying before you set a budget.
Here is the strategic read in one breath. Owned channels have a real up-front cost and a slow start, then their cost per lead falls toward zero as they compound. Rented but exclusive channels start fast and stay flat, and they are honest about it. Shared lists start cheap and stay bad, and they never build you anything. So fund the fast exclusive channel now to keep the calendar full, build the owned ones for the years ahead, and let the shared list go the day the phone can survive without it.