5 Lead Generation Metrics Every Mortgage Lender Gets Wrong

In 2000, I joined DeepGreen Bank as Employee #7 and helped launch the industry's first unconditional online HELOC. In 2002, I went to Quicken Loans as VP of National Home Equity and built EquityOnline. In 2005, I founded Kaleidico, and I've been running lead generation programs for mortgage lenders ever since.

Across that 25-year stretch of building, buying, and converting mortgage leads, there are five lead generation metrics I see mortgage lenders measure wrong almost universally. These aren't small errors. Each one materially distorts operational decision-making and usually costs the lender money in ways they never surface.

Here are the five, and what I use instead.

1. Cost per lead

The most-tracked metric in mortgage lead generation is also the least informative.

Cost per lead tells you what you paid. It tells you nothing about what you got. A lead from Vendor A at $40 and a lead from Vendor B at $80 look like a 2x price difference on the CPL report. In my experience, the actual economic difference between the two is almost never 2x and is frequently inverted — the $80 lead often produces more revenue per funded loan, a higher contact rate, and a shorter payback period than the $40 lead.

CPL is an input variable. It should not be the headline number on any lead gen dashboard.

What I watch instead: Effective CPA per funded loan, by source. The actual question is, how many dollars did I spend in this channel, divided by how many dollars of funded revenue did I generate from this channel? That's the number that tells me whether the channel is working. Every other lead-gen metric is supporting detail.

2. Conversion rate

The second most-cited metric after CPL is conversion rate, usually stated in the abstract ('our conversion rate is 8%') as if that number were comparable across periods or across lenders.

It isn't. Conversion rate is a ratio between two variables — the numerator (funded loans, or applications, or whatever stage you measure) and the denominator (leads bought, or applications taken, or whatever precedes the numerator). Both the numerator and the denominator can be defined differently by different operators, different periods, and different vendors. An aggregate conversion rate of 8% at Lender A is often not comparable to 8% at Lender B.

Even within the same lender, the aggregate conversion rate hides the variance that actually matters. A lender converting 10% on good vendors and 2% on bad vendors has a meaningfully different operational picture than a lender converting a uniform 6% across all sources.

What I watch instead: Cohorted conversion rates, segmented by lead source, product mix, and time vintage. The aggregate number is the headline for the CFO meeting. The cohorted view is what actually drives operational decisions.

3. Lead volume

Every lender I've ever worked with has been able to tell me how many leads they bought or generated last month. Most of them have not been able to tell me what percentage of those leads were ever contacted by a human, and most of the remainder can't tell me what the distribution of contact attempts per lead looks like.

Lead volume without a quality-of-work overlay is a vanity metric. It feels important because it's a big number, but it tells you almost nothing about the productive output of the program.

I've audited lenders buying 5,000 leads a month where 40% never received a single dial attempt. The lead cost spent on those 2,000 leads was effectively zero-return. The lender tracked the 5,000 lead count in their dashboard and treated it as a sign of program scale. It was actually a sign of program waste.

What I watch instead: Worked leads — specifically, leads that received the full dial cadence (typically 8-12 attempts over 14 days). That's the number of leads your operation actually worked. If worked leads is dramatically lower than purchased leads, you have a capacity or process problem, and buying more leads won't fix it.

4. Speed to first response

Most lenders I meet track this metric, and most of them track it poorly.

The common mistake is measuring speed to first response as an aggregate, with any response counted — an automated email, an SMS autoresponder, a voicemail. Those are not first responses. The lead doesn't care that your CRM fired an auto-email 12 seconds after form submission. The lead cares whether a human loan officer actually got them on the phone.

The metric that moves conversions is time from lead receipt to live conversation with a human. Everything else is theater. I've watched lenders celebrate sub-minute 'response times' that were purely automated, while their actual human contact time was measured in hours.

What I watch instead: Median time from lead receipt to live human conversation, on leads where contact was eventually made. Broken out by hour of day. Broken out by day of week. This metric surfaces the capacity gaps and the staffing misalignments that are costing the operation real money.

5. Average loan size

Mortgage lenders love to track average loan size because it's an easy number to produce and it usually trends in a direction they can spin as positive. It is also nearly useless as a standalone metric.

Average loan size is driven by geographic mix, product mix, market conditions, and the distribution of borrowers a lender happens to attract — all variables that have very little to do with the lender's operational performance. A rising average loan size can mean the lender is winning more affluent borrowers. It can also mean the lender is losing the lower-loan-size business, concentrating into jumbo products, or simply being carried along by housing price appreciation.

When a lender cites rising average loan size as a positive KPI, I almost always discover that the underlying portfolio concentration is getting worse, not better.

What I watch instead: Loan size distribution — what percentage of funded loans falls into each $100K bucket. That view shows whether the business is actually diversified or is concentrating into a narrower slice of the market. Concentration looks like strength in a good rate environment and like fragility when the environment shifts.

The pattern underneath the five mistakes

Look at these five metrics together and a pattern emerges. The commonly-tracked mortgage lead generation metrics are almost all aggregate, absolute, or volume-oriented. The metrics that actually drive operational decisions are cohort-based, ratio-based, or quality-adjusted.

That's not a coincidence. Aggregate metrics are easier to produce. They're what the LOS and CRM report by default. They make for good executive dashboards. Cohort-based metrics require more effort, more discipline, and often some custom data work.

The lenders that win the next five years will be the ones that put in the work to measure the right things. The ones that keep celebrating monthly volume and average loan size while their effective CPA rises and their margin compresses are the ones that are going to show up in the 2028 consolidation stories.

A practical starting point

If you're a mortgage lender running on the standard KPI set and you want to upgrade, here's the first step I recommend.

Produce one view, updated weekly, that shows these four numbers by acquisition channel:

1. Dollars spent

2. Funded loans produced

3. Effective CPA (dollars spent divided by funded loans)

4. Median time to live human contact for leads from that channel

That's it. Four numbers, weekly, by channel. If you can't produce this view today, getting to where you can produce it is the single highest-leverage KPI work you can do.

Once you have the view, run your operations meetings off it. Kill channels where the effective CPA exceeds your margin envelope. Double-down on channels where it's below. Fix the speed-to-contact issues on channels where the response times are clearly inadequate.

The compounding impact of running a tight program on the right metrics is, in my experience, the single biggest operational advantage available to a mortgage lender right now. It's harder to build than a new marketing campaign, but the return on the effort is substantially larger.

I help mortgage lenders build this kind of measurement discipline through Kaleidico and advise on lead acquisition economics more broadly at Bill Rice Strategy Group. The full framework for the buy side of this is in The Lead Buyer's Playbook.