Angi math: what a booked job actually costs you

Ask whether Angi Leads is worth it in any contractor forum and you get a fight. One guy built his whole schedule on it. Another canceled after three months and still gets the sales calls. Both are telling the truth, because "worth it" hides three numbers, and most owners never multiply them together.

The numbers: what you pay per lead, how many of those leads you actually reach, and how many of the reached ones you close. Multiply them and you get your real cost per booked job. That's the only number that settles the argument, and only your own books can produce it.

Give Angi its due first

The old complaint, "they blast my number out on junk shared leads," is dated. Angi changed the core mechanic: since January 13, 2025, nearly every homeowner who submits a request actively chooses which pros to contact, per Angi's own Q4 2024 shareholder letter. The auto-distribution era is all but over (Angi reserves a separate add-on product for non-chosen network leads). If a lead reaches you now, odds are the homeowner picked you, and that's a genuinely better lead than the old model produced.

What didn't change is the shape of the deal. You pay per lead, not per job. Angi doesn't publish a rate card, but third-party trackers put lead fees around $15 to $85, pushing $100 in some markets and trades, on top of a membership fee of roughly $300 a year. Treat those as ranges. Your invoice is the truth.

The record worth knowing

There's also a documented reason contractors don't take lead-quality claims on faith. In March 2022 the FTC charged HomeAdvisor, Inc., which also did business as Angi Leads, with making false, misleading, or unsubstantiated claims about its leads since at least mid-2014: leads pitched as matching a pro's trade and area that often didn't, leads bought from affiliates rather than coming from people who'd sought out HomeAdvisor, and job-conversion rates it couldn't substantiate. The settlement, proposed in January 2023 and finalized that April, required payments of up to $7.2 million and barred the deceptive pitches.

That conduct predates the homeowner-choice era. The lesson survives it: never take a conversion rate from a salesperson. Compute your own.

The formula

Here it is:

Cost per booked job = lead fee ÷ (contact rate × close rate)

Three inputs, all from your own numbers:

• Lead fee. Your average from the last Angi invoice, after any credits you actually won back. Not the advertised floor.
• Contact rate. Pull your last 20 or 30 paid leads. How many did you get on the phone at all? Ghosts still cost the full fee.
• Close rate. Of the people you reached, how many turned into a completed, paid job?

Say a lead runs $60, you reach 6 in 10, and you close 1 in 3 of those. That's $60 ÷ (0.6 × 0.33), about $300 per booked job. Every figure in the table below is illustrative. Plug in yours.

Lead fee	Contact rate	Close rate	Cost per booked job
$40	70%	50%	$114
$60	60%	33%	$303
$85	40%	25%	$850

Now hold that against the job. $850 to win a $250 drain call is a losing trade every single time. $303 to win a water heater install that bills a few thousand might be a perfectly good one. Same platform, same fees, opposite verdicts, which is why "is Angi worth it" has no general answer. It has your answer, per trade, per job type.

One warning about stale inputs: if your close rate memory comes from 2023 shared leads, remeasure. Homeowner-choice leads should close better. Give the platform the credit your own data supports, and nothing more.

What owners actually report

For flavor of the inputs, here's a 2024 thread on Mike Holt's electricians forum, where a contractor asked whether pay-per-lead sites are worth it. One member, Ponchik, on the customers the leads brought:

I would get tire kickers and "over the phone ball park pricing" customers.

Another, LarryFine, on getting bad leads refunded:

They come up with excuses to justify not crediting for bad leads.

Two members of an electricians' forum, in one thread posted while the homeowner-choice rollout was still underway. Not a survey, and your books outrank both of them. But "tire kickers" and "fight for credits" are exactly the contact rate and the effective lead fee in the formula above, told as stories.

Renting demand, now inside ChatGPT too

Even when your math clears, look at what the fee buys: a chance at a customer who lives in someone else's system. The homeowner's account is an Angi account. The review lands on your Angi profile. Next project, they open the app, and you pay again.

That rent is now following customers into AI assistants. In March 2026, Angi launched its app inside ChatGPT: a homeowner asks a home-repair question, gets matched with pros, and clicks through to Angi to hire. Discovery moving into chat didn't remove the toll booth. It relocated it to the new road.

Our opinion, plainly: paying for discovery can be rational, especially while you're filling a schedule. Paying rent on customers you already earned never is.

Own the surface you already have

To be clear about what we do and don't do: ServiceFront won't get you found on Angi, Google, or ChatGPT. Discovery stays your problem, however you solve it. What we handle is the demand that already arrives at your own site: the referral, the repeat customer, the homeowner comparing quotes at 9pm. Today most of that traffic hits a dead contact form and leaves. ServiceFront puts an agent there instead, one that answers questions from facts you've verified (no invented prices, ever), flags the emergencies, and captures the job details, then emails you the lead. No per-lead fee on demand that was already yours.

If you want to see it against your own business, paste your URL into servicefront.ai and ask it the questions your customers ask you.