What a mortgage lead looks like: credit, home value, income

A mortgage lead we delivered over the 30 days to 26 July 2026 averaged a 695 credit score, a $583K home value and $74K of household income. That is our own production data rather than an industry estimate, and the rest of this page is the caveats that go with it.

Source: VisionXLab production delivery data, the 30 days to 26 July 2026. First-party measurement, stated as such. No outside party has audited it, which is the same caveat we apply to everybody else's numbers on this site.

Avg credit score

695 Reported by the borrower, screened before delivery

Avg home value

$583K The borrower's own estimate, not an appraisal

Avg household income

$74K Only the products our form asks it on

What these three numbers are, and what they are not

They are self reported. A borrower types a credit band, a property value and an income into our form, and our qualification screen holds them to those answers before anything is delivered. There is no credit bureau pull anywhere in that path. That is the whole method, and it is also the honest limit of it: what we can tell you is what a borrower said about themselves and that the answer cleared our screen, not what a bureau file says.

They understate rather than flatter. Our brackets are open ended at the top, and every answer in a top bracket is counted at that bracket's lower bound. A borrower who picks the highest home value band is arithmetically identical to one sitting exactly on its floor. So the true averages are above the printed ones, by an amount the data cannot tell us.

They are a snapshot, not a feed. The window closed on 26 July 2026 and every figure here moves when it is taken again. Nothing on this page reads a database when you load it, and any page claiming otherwise is worth asking how.

The single average is the least useful cut of it

Four products, four different borrowers. A cash-out borrower and a purchase borrower have almost nothing in common except a credit score that lands in the same decade. Each product below shows only the fields our qualification screen asks on it, which is why the shorter lists are short.

Source: VisionXLab production delivery data, the 30 days to 26 July 2026. Averages over delivered leads only. Equity and loan to value are computed from the borrower's own stated home value and mortgage balance, so they inherit every caveat above.

The loan to value lines are the ones worth sitting with, on Refinance and Cash-Out Refinance above, because loan to value is what decides whether a file is workable before anything else about it matters. Ours are arithmetic on two numbers a borrower typed, which is a weaker thing than a lender's own calculation and should be treated as one.

What does not reach a client

Not everything our advertising produces reaches a client, and the gap is the part a lead vendor never shows you. Over the same window, this is what happened to the leads in the sample.

Read that table with one caveat attached. The shares come from a sample of leads drawn from the same window rather than from the whole of it, and we do not publish the size of that sample. They describe the shape of the window and they are not a population measurement. The averages above them, and the state shares below, are computed over the whole window.

Every lead that failed the screen carries the rule that stopped it. Grouped by the field that failed:

Credit is 98.1 percent of screen failures on its own. That is not a boast, it is what a credit floor is for, and it is the thing to hold on to when reading the average at the top of this page: 695 is a post-screen figure, and the population it was drawn from is not the population it describes.

Where the volume is

Share of delivered volume by state, largest first.

Those add to 58 percent of delivered volume over the window. Each share is rounded to the nearest point, so the total carries that rounding. The remainder is spread nationwide. This is a map of where our own advertising ran and where clients hold licenses, not a map of the United States mortgage market, and it moves when either of those moves.

How much weight to put on this

Read it the way we ask you to read everybody else's numbers. It is first party, which means the company with an interest in the answer is also the one holding the measuring instrument. Nobody outside has audited it. It covers thirty days rather than a year, and thirty days of our own advertising rather than of the market. Where a figure like this sits against the industry's published ones, and which of those have a traceable source at all, is in what the data actually supports between a lead and a funded loan.

What it is good for is the shape. Once a screen has been applied, the borrower behind an internet mortgage lead averages 695 on credit and $74K on household income, and 36 percent of what the screen sees never gets past it. If a vendor quotes you a profile with no window, no method and no note on who reported the fields, those three absences are the finding.

The short version

• 695 average credit score, $583K average home value, $74K average household income, on leads delivered over the 30 days to 26 July 2026.
• All three are reported by the borrower and screened against those same answers. No bureau pull.
• Open ended top brackets count at their lower bound, so all three understate.
• 36 percent of a sample from the same window failed the screen and was never delivered. Credit was 98.1 percent of those failures.
• CA alone was 34% of delivered volume. The rest is spread nationwide.

The lead by lead view of the same window, contact details stripped, is not published. It gets walked through line by line on a call. Buying questions rather than research ones are on our FAQ, and the operating half of this, how fast and how often a desk has to dial what it receives, is in how many calls it takes to reach a mortgage lead.