I want to talk about something that almost everyone in this industry pays for and almost nobody admits doesn’t work.
Contractor lead lists.
You know the ones. A vendor reaches out (or you fill out a form on their site). They sell you a CSV of 1,200 “qualified construction leads in your zip code” for somewhere between $0.40 and $2 a record. You import them into your CRM. Your team starts dialing. About a third of the numbers ring through to nothing or someone unrelated. Another third reach a homeowner who has no idea what permit was pulled at their address and is annoyed you’re calling. The remaining third—if you’re lucky—contains five real opportunities.
The math, when you actually run it, is brutal.
Why the lead-list model is broken
There are three structural problems with the lead-list business that no amount of CSV polish can fix.
1. The contact data is bad on day one.
Lead-list vendors aggregate from dozens of upstream sources—skip traces, scraping, public records. The phone numbers on those records are accurate, on average, about 60% of the time. The email addresses are accurate about 40% of the time. By month three, both numbers drop another 10–15 points as people move and change jobs.
And here’s the thing nobody tells you: even the contacts that are correct are usually the wrong person. The phone number on a building permit is often the architect, the engineer, or the project administrator—not the decision-maker who hires GCs. You’re paying for accurate contact data for the person who can’t hire you.
2. The records are not contextual.
A lead-list record is a single row: address, project type, valuation, applicant name. What it doesn’t tell you: where this project is in the lifecycle, whether the GC has already been chosen, whether the developer has done four other projects with a competitor of yours, whether the parcel was just rezoned, whether the plat was filed last week or fourteen months ago.
All of that context is the actual signal. Without it, you’re calling cold into a project where the decision was made six months ago.
3. Everyone else has the same list.
This is the worst part. Lead-list vendors sell the same list to dozens of contractors. The day you call your “new” lead, you’re the seventh person who’s called them this week, and the third person from your trade. The homeowner is annoyed before you even introduce yourself.
We used to be on lead lists too. We stopped about eighteen months ago. Here’s what replaced them.
The playbook that replaced lead lists
The thing that finally worked is a categorical shift, not a tactical one. We stopped buying contact data and started watching the project record itself.
- 01 ·Watch the project, not the contact. Every active project in your market is publicly trackable—the plat, the plan review, the permit, the inspections. We started monitoring the project lifecycle and only doing outreach at moments where outreach actually makes sense.
- 02 ·Time the outreach to the lifecycle stage. A plat hits Planning Commission—wrong moment to call. The plan review just cleared structural and the developer is shortlisting GCs—right moment. The cold-call success rate jumps 4–6x when you time it correctly.
- 03 ·Reach the LLC, not the address. The owner of a project isn’t at the project address; they’re at the LLC mailing address (often a CPA or a registered agent). Pull the LLC, find the principals on the Secretary of State, and reach the actual decision-maker.
- 04 ·Run the same project through multiple touchpoints. The architect, the engineer, the developer, and the parcel owner are often four different people. Hitting two or three of them in the same week is dramatically more effective than hitting one of them once.
What the numbers look like
We tracked the change carefully because we wanted to know whether this was real or just a story we were telling ourselves. Twelve months in:
- Cold-call connect rates went from about 12% (lead lists) to 41% (project-driven, time-boxed outreach).
- Conversations-per-week roughly held steady (about 30 per rep).
- Bid invitations went up about 3.2x—we were being invited to projects that lead-list outreach never put us in front of.
- Cost per opportunity went down by about 70%, mostly because we stopped paying $1,800 a month for lists that produced four real conversations.
The number that mattered most, though, is one we couldn’t measure cleanly: how the team felt about the work. Nobody enjoys cold-calling a list of strangers who didn’t ask to hear from you. People do enjoy reaching out about a specific project they have something useful to say about. Morale is part of the math, and the lead-list model hides the cost of it.
What about the AI lead generators?
Reasonable question. There are now a handful of products marketing themselves as “AI lead generators for contractors.” Some of them are real. Most of them are repackaged lead lists with a chatbot.
The honest test: ask any AI lead vendor to show you the lifecycle of a single project across the public record—from plat to permit. If they can’t, you’re looking at a CSV with a friendlier UI. If they can, you’re looking at something genuinely different.
I wrote a longer breakdown of how to evaluate AI lead intelligence a few weeks back if you’re actively shopping. Quick version: the right tool isn’t a faster way to dial bad numbers; it’s a way to dial fewer numbers, better-timed.
If you’re still on lead lists
I get it. They’re cheap, they’re predictable, they fit nicely into a CRM workflow your team already uses. The switching cost feels real.
Try one experiment for a quarter. Cancel one of your lead-list subscriptions. Take that budget and put it into one of two things: either a tool that actually watches the project record (we’d argue that’s us, but there are alternatives), or a part-time analyst whose job it is to read public agendas and brief your sales team weekly. Run the experiment for 90 days. Track it the way I described above.
If you go back to lead lists at the end, fair enough. We didn’t.