Why so many agents decide internet leads don't work, the numbers to know before you judge a source, and the follow-up that makes them pay.
Before you write off a lead source, here's what I'd check first. Most "bad lead" verdicts don't survive these:
Know your cost per bind against your carrier's comp. Not a feel for it, the actual number.
Know your waterfall: contact rate, quote rate, bind rate. A publisher who's worked your carrier can back those out for you.
Hold a cadence that runs 45 to 60 days, and use the full 90-day TCPA window. Most agents quit in week one.
Be first to contact, ideally inside the first 30 seconds. Speed is the cheapest edge in this business.
Match your volume to what your team can actually work. More leads than you can work is just a bigger pile to feel guilty about.
This got sharper than it was a few years back because the economics moved. A call that ran around $35 in 2016 is $150 to $175 now. On the captive side, carriers are consolidating their tech from corporate down. Across the board the outreach model keeps cycling, leads, then calls, now some hybrid, and plenty of agencies are hiring third-party teams offshore to chase a cheaper touch. The consumer moved too. People re-shop constantly now; someone buys a policy and is back in the market in 60 days hunting a lower rate. And TCPA and FTC rules keep tightening underneath all of it. So the room for a sloppy process to still turn a profit has gotten small. When leads were cheap, waste hid. At today's prices, it doesn't. That pressure is exactly why so many agencies' lead-buying strategies are quietly failing, and it's fixable.
I hear the same line at every conference. "I bought a hundred leads, worked them a few days, closed two. The vendor sold me junk." I get it. Plenty of you have been burned, and I won't talk you out of something you actually lived. But I ran the Accelerated Growth Program at EverQuote, the arm that consulted agencies on scaling with these leads, and I've sat inside more than a thousand agencies watching how the work really gets done. The thing that stuck with me: two shops would buy the same leads, same source, same week, and one built a book while the other swore they were worthless. When the leads are identical and the outcomes aren't, the leads were never the variable.
So here's the question I ask every agent who tells me leads don't work: what's a good lead to you? Most can't answer it. They're grading on a feeling. Get specific instead. Know the cost per bind that keeps you profitable against your carrier's comp, because what works for an Allstate captive is different from an independent writing ten carriers. Then a batch of leads isn't good or bad in the abstract. It either clears your number or it doesn't.
The part that actually decides it is what happens after the lead lands. The top agencies are first to contact, usually inside 30 seconds, and they run a cadence 45 to 60 days deep, using the full 90-day window to bring a consumer back to market. The ideal customer, a multi-car homeowner with good credit, is working nine to five same as you. There's no infinite grid of them, so when one lands, you stay on them. But be honest about what a real cadence on a hundred leads looks like: hundreds of dials, run the same on lead 1 and lead 100. Nobody holds that on willpower. By Thursday the energy dips, calls slip, and a couple of leads that would have bound never get worked. Then the leads take the blame for a cadence that quietly fell apart.
I see the bigger version of this too. Agencies get so focused on feeding the machine, more marketing dollars, more volume, chasing the comp plan, that they lose sight of their margin and their cost per acquisition. When I built agencies up through AGP, I wouldn't let them lean hard into buying leads until three things were true: a real process everyone runs the same way, the systems to execute it consistently so it doesn't live in one producer's head, and enough capacity to work the volume they were about to buy. Process, systems, size, then volume. Skip that order and no lead source on earth saves you. If you want the version of this written as a straight playbook, our CGO laid out how to actually make purchased leads work, and it lines up with everything I've seen from the field.
"Systems and size" used to mean hiring and managing more bodies, which is its own hard problem right now. That's the part AI changes. Something like Mav holds the cadence the same way on lead 1 and lead 100, works the full follow-up window, and puts a live, interested person in front of your producer. You hit the numbers without standing up a call center to do it. The process still has to be yours. The discipline to run it just doesn't have to come out of one person's willpower anymore. And with TCPA and FTC both moving, having consent and opt-outs handled inside the workflow instead of in someone's memory is its own quiet advantage.
So before you decide the leads are the problem, one question: can you say your target cost per bind out loud, right now? If you can't, that's where I'd start, not with a new lead source. The leads were only ever raw material. What you do in the 60 days after they land is the whole game.
If you're stuck on that question, tell me where, is it knowing the number, or holding the process long enough to hit it? That's the conversation I have most, and what I hear back is usually the real bottleneck. All said, the agencies that pull ahead got clear on their numbers and built the process to earn them. Better leads were never the thing.
Are internet leads even worth it anymore?
They're worth it if you have the process to work them and dead weight if you don't. I've watched the same leads make one agency and frustrate another. What decides it is your speed, your cadence, and your discipline over a 60-day window. Buy what your team can actually work, and they're some of the best growth money you can spend.
What should my cost per acquisition be?
There's no single right answer, and anyone who hands you one without asking about your carrier is guessing. It comes down to your comp and what a bind is worth to you. Figure out the cost per bind that keeps you profitable, then ask your lead partner to back out the contact, quote, and bind rates for your carrier so you can see whether a source clears that bar. Know that number before you spend a dollar.
How long should I follow up before I move on?
Longer than most agents do. The best shops run a cadence 45 to 60 days deep, and you've got a 90-day TCPA window to bring someone back to market. Most quit in the first week, right when the consumer was getting around to dealing with it. The bind is often sitting in week three, waiting on the agent who's still there.
