Most mortgage brokers who buy leads can tell you exactly how much they paid per lead last month. Very few can tell you what happened after those leads landed on their website.
That’s the problem this article is about. Not lead quality, not lead sources, not which vendor charges what. The problem is simpler and more expensive than all of those: brokers keep pouring money into the top of a funnel that leaks from the middle.
The Math Brokers Already Know (and the Math They Don’t)
The lead-cost side of the equation is well documented across industry reporting. The average cost per mortgage lead in 2026 ranges from roughly $15 to $75, depending on lead type and exclusivity. Shared leads (sold to multiple lenders at once) run $15 to $30. Exclusive leads, where only one broker receives the contact, run $40 to $75. At the higher end, some exclusive purchase leads can hit $100 or more. And those numbers are climbing. HousingWire reported in early 2026 that internet lead pricing on major aggregator platforms has risen roughly 45% year over year, driven partly by the trigger-lead ban forcing lenders toward costlier, high-intent channels.
Most brokers track those numbers closely. What they track far less often is what happens next.
A broker spends $3,000 on 100 leads at $30 each. Industry-reported benchmarks put shared lead conversion at roughly 1 to 3 percent. That means 1 to 3 of those 100 leads will close. The other 97 to 99 went nowhere. That’s $1,000 to $3,000 in lead costs per funded deal, before staff time, follow-up, or any other overhead. The difference between closing 1 and closing 3 is not about getting better leads. It’s about what happens between the click and the closed loan.
Now look at the other side. The Chrisman Commentary notes that in a typical retail mortgage environment, conversion from application start to completed submission hovers around 60 to 70 percent, meaning 30 to 40 percent of borrowers who start never finish submitting.
Lead cost ranges are reported across multiple lead marketplaces and are consistent across sources, though no single independent body publishes an official benchmark. The application completion figure is specific to retail mortgage environments as of 2026. The exact rates at your brokerage will depend on your own lead sources and application flow, but the direction is clear: a meaningful share of started applications never get submitted, and that gap sits between the click and the closed loan, not before it.
Where the Funnel Actually Leaks
When a purchased lead clicks through to a broker’s website, what happens? In most cases, one of three things:
They hit a long application form immediately. The visitor just arrived. They may have been browsing rates or comparing options. Now they’re being asked for income, employment history, SSN, and a stack of documentation. Many aren’t ready for that. They were expecting a quick answer, not a 20-minute commitment. So they leave.
They find a calculator but nothing after it. The calculator gives them a payment number. That’s useful. But it doesn’t tell them whether they’re in a position to actually get approved. And after the calculation, there’s no next step, no prompt, no path forward. They got a number and they’re on their own with it.
They see a contact form and nothing else. No calculator, no estimate, no sense of where they stand. Just a form asking for their name and phone number with a promise that someone will call back. For a shared lead that’s already going to three or four other brokers at the same moment, giving up contact information with no value in return is a hard ask. Most won’t.
None of these are unusual setups. They’re the norm. And they’re the reason more leads don’t fix the conversion problem. The leads aren’t necessarily the problem. What’s waiting for them on the other end often is.
Why Buying More Leads Makes This Worse
This is the part that feels counterintuitive but is simple math once you see it.
If your website converts 2% of visitors into real conversations, buying 100 more leads gives you 2 more conversations. Buying 200 more gives you 4. You’re scaling linearly against a fixed, low conversion rate. Every additional lead costs the same, but the funnel still loses the same percentage.
Meanwhile, the cost of those leads is rising. Exclusive leads run $40 to $75 each in 2026 according to industry reporting, and HousingWire noted a 45% year-over-year increase on major platforms. As more brokers compete for the same lead pools, prices go up. A broker spending $3,000 a month on shared leads with a 2% site conversion rate is closing roughly 2 deals. Doubling the lead budget to $6,000 gets roughly 4, at the same cost per deal. Nothing improved except the total spend.
Compare that to fixing the conversion rate. If the same broker keeps the $3,000 lead budget but improves site conversion from 2% to 4%, they close 4 deals instead of 2, with zero additional lead spend. The cost per funded deal drops by half.
| Approach | Monthly lead spend | Leads | Site conversion | Funded deals | Cost per deal |
|---|---|---|---|---|---|
| Current | $3,000 | 100 | 2% | 2 | $1,500 |
| Buy more leads | $6,000 | 200 | 2% | 4 | $1,500 |
| Fix the funnel | $3,000 | 100 | 4% | 4 | $750 |
The numbers in that table are illustrative, not drawn from a specific brokerage. “Site conversion” here means the full path from a lead landing on the site to a funded loan, which includes steps outside the website’s control (underwriting, appraisal, borrower decisions). The website’s direct influence is on the earlier part of that path: whether a visitor stays, engages, and moves toward an application at all. That’s the part this article is about. But the relationship holds regardless of where you draw the line: improving the conversion rate of the leads you already have is almost always cheaper than buying more leads at the same conversion rate.
Fixing the Middle Before Filling the Top
If the funnel leaks in the middle, the fix has to happen in the middle. Not at the top (more leads), not at the bottom (better closers), but in the space between a visitor landing on your site and deciding whether to take the next step.
That space is where a visitor needs to go from “I’m browsing” to “I know where I stand.” A calculator alone doesn’t get them there, because it answers what their payment would be, not whether they’re in a position to get approved. A raw application form doesn’t get them there either, because it asks for commitment before offering any clarity.
The sites that handle this well tend to do one thing the rest don’t: they give the visitor something useful about their own situation before asking for anything in return. That might be a readiness signal, an estimated approval range, a DTI check, or a combination of those. The specifics matter less than the order. Value first, then the ask.
A visitor who knows their DTI, readiness score, or approval likelihood has a reason to take the next step. A visitor staring at a blank form does not.
When a contact form follows a readiness check, it feels connected to something the visitor already cares about. When it shows up cold, it reads as a random ask.
Every visitor who bounces from your site without understanding where they stand is a lead you already paid for and got nothing from. Fixing that costs less than buying more.
This is the gap I built MDE Pro to fill. It sits between the calculator and the application, giving a visitor a readiness score and approval likelihood in about 20 seconds before they continue into the broker’s own application system. No data is stored, no leads are captured on our end, and the broker’s name and NMLS number stay visible the entire time. It’s one way to close this specific gap. It’s not the only way, but the gap itself is real whether you use this tool or not.
See What Fixing the Middle Looks Like
MDE Pro gives visitors a readiness score and approval likelihood before they reach your application, on your site, under your name.
Try the MDE Pro DemoMore leads through the same leaky funnel gives you more leaks, not more loans. Fix the middle first.
