Stop Renting Leads from Angi, HomeAdvisor & Thumbtack
Why aggregator leads are a poor foundation for contractors, the unit economics most owners skip, and the channels that compound into a real lead engine.
Almost every contractor I work with has the same complicated relationship with Angi (formerly Angie’s List), HomeAdvisor, and Thumbtack: they hate paying for the leads, they hate competing against three other contractors for the same lead, they hate the price-shopper customer mix — and they keep paying because cutting it off feels like deliberately starving the business.
The problem with that math is the budget you’re feeding into aggregator leads is almost always the same budget that would, if redirected over 6–12 months, build a lead engine that delivers more leads at lower cost with higher close rates.
This article makes the unit-economics case, walks through the alternatives, and lays out the transition plan we use with clients.
This sits in the Send chapter of Why Your Website Isn’t Ringing.
The actual unit economics
Realistic 2025 numbers from contractor accounts we’ve audited:
| Metric | Aggregator leads | Map Pack leads (organic) | Local Service Ads |
|---|---|---|---|
| Cost per lead | $30–$200 | $0–$15 | $20–$100 |
| Lead exclusivity | Shared with 3–5 contractors | Exclusive | Exclusive |
| Close rate | 8–18% | 25–40% | 30–45% |
| Cost per booked job | $200–$1,200 | $0–$60 | $50–$300 |
| Customer mix | Price-shoppers | Mixed | Buyer-intent |
| Long-term ownership | None — leads belong to the platform | Full | Full |
The summary: aggregator leads cost 4–10x more per booked job than your own channels, on lower close rates and worse customer mix. They look reasonable when you only count the per-lead cost — they look brutal when you run the booked-job math.
A real example from a residential plumbing client we audited in early 2025:
- Aggregator spend (Angi + HomeAdvisor): $4,400/month
- Leads received: 92/month average
- Booked jobs: 11 (12% close rate on shared leads)
- Cost per booked job: $400
- Average ticket: $480
- Net contribution per booked job: $80 (20% gross margin × $480 = $96 minus $400 lead cost = -$304 margin loss per job before any operating cost)
That client was losing money on Angi and didn’t realize it because they were averaging the cost-per-lead across all aggregator activity instead of looking at the booked-job economics.
Why the aggregator model is structurally broken
Three reasons aggregator leads underperform on close rate:
1. Shared leads create a price war
When the same lead goes to 3–5 contractors, the customer becomes a price-comparator by default. They expect 3 quotes. They expect to negotiate. The contractor who wins is usually the cheapest, which creates a race to the bottom that erodes margin even on the leads you do close.
2. The aggregator owns the customer relationship
When a customer “wins” a contractor through Angi, they associate the value with Angi, not with you. Re-engagement, repeat business, and referrals flow back to the platform, not to your shop. You’re effectively renting the customer for a single transaction.
3. The intent quality is mixed
Aggregator forms tend to attract a higher proportion of low-budget shoppers, “research mode” shoppers, and tire-kickers than direct Google search does. Your time gets eaten by quotes you were never going to win.
When aggregator leads do make sense
I’m not arguing for cutting them off entirely. There are use cases:
- Slow-season filler. If your November pipeline is thin and your fundamentals don’t yet produce enough volume, aggregators can plug the gap.
- New geographic markets. Entering a new town or service area where you haven’t yet built Map Pack or organic presence.
- Specific service categories you don’t market on your own site. A plumber who doesn’t market water-treatment systems but takes them when they come in might use Angi to source them.
In each of these cases, aggregators are a tactical tool, not a strategic foundation.
The alternatives that compound
Three channels that, over 6–12 months, replace aggregator volume at lower cost:
1. Map Pack and organic search
The fundamentals from Local SEO for HVAC, Plumbing & Roofing — The 2026 Playbook and How to Rank in the Google Map Pack for Service-Area Businesses. Slow to build, durable once built, near-zero marginal cost.
A shop that holds top 3 in the Map Pack for its primary service in its primary town is generating 60–250 inbound leads/month at zero per-lead cost. The same volume from Angi would cost $1,800–$50,000/month.
2. Local Service Ads (where available)
LSAs are pay-per-lead like aggregators but with three big differences:
- Exclusive leads (not shared)
- Google verifies you (license, insurance, background check)
- Better lead quality (Google’s intent matching is sharper than aggregator forms)
Cost per lead is comparable to aggregators, but close rates run 2–3x higher because of the exclusivity. For HVAC and plumbing in most markets, LSAs are the clearest “yes” in the paid acquisition space.
3. Referral and repeat customer systems
The cheapest leads you can possibly get are referrals from past customers and repeat business from your existing list. Most contractors do almost nothing systematic here.
A working referral system:
- SMS to every customer 30 days post-job: “Hope everything’s still working great. We rely on referrals — anyone you know who needs [service]? We’ll take great care of them.”
- Annual maintenance reminder to every past customer
- Spring/fall reactivation campaign to customers who haven’t booked in 12+ months
This produces 10–20% of total inbound volume for shops that run it consistently, at essentially zero acquisition cost.
The transition plan
If your aggregator spend is currently 40%+ of your lead budget and you want to taper it down without breaking the business, here’s the 12-month plan we run with clients:
Months 1–3: Build the foundations
- Map Pack optimization (categories, reviews, posts, services)
- Service-area landing pages for top 3–5 towns
- Missed-call text-back live
- Standard follow-up sequence live
- LSAs launched (where available)
- Referral SMS automation live
Aggregator spend stays flat through this period. Don’t cut anything yet.
Months 4–6: Measure and rebalance
- Map Pack starts producing meaningful volume
- LSAs settle into their cost-per-lead range
- Referral volume starts building
- Begin reducing aggregator budget by 15–20% per month, redeploying half of the savings into stronger LSA bidding and content investment
Months 7–12: Lean into the compounding channels
- Map Pack rankings consolidate into top 3 in priority towns
- Organic search starts driving meaningful inquiry volume
- Referrals plus repeat business hit 15–25% of total lead mix
- Aggregator spend drops to 5–15% of lead budget, used as slow-week filler only
By month 12, most clients running this plan have:
- 2–3x more total inbound leads than at month 0
- 50–70% lower aggregator spend
- 30–60% lower blended cost per booked job
- Higher gross margin per job (because they’re no longer competing against 3 other quotes from shared leads)
What this requires from you
The plan only works if you commit to the foundational work. The contractors who fail at this transition are usually the ones who:
- Try to cut aggregator spend to zero in month 2 before the replacement channels are built
- Don’t follow through on the GBP optimization (no weekly posts, no review automation, no owner replies)
- Treat content as optional and never build out the service-area page network
- Skip the missed-call text-back and follow-up sequence, so even the leads they do generate from new channels die at the back of the funnel
Read The Contractor Follow-Up Sequence (Day 0 → Day 30) for the back-of-funnel piece, and Trust Signals: Reviews, Licenses & Photos That Make Visitors Call for the on-site work that supports new lead conversion.
The full Send framework is in Why Your Website Isn’t Ringing. If you want help building the transition plan for your specific business, book a 15-minute call and we’ll walk through your current numbers.
You don’t need to keep paying $400 per booked job. There’s a better way and it doesn’t take a year of misery to get there — but it does take 6–12 months of patient compounding.