If you’ve spent any time around the mortgage industry’s online discourse, you’ve probably seen the recapture rate stat thrown around: somewhere between 18% and 22% of borrowers refinance or repeat with the same LO who handled their original loan. The number is consistent across major studies, and it’s been roughly the same for the better part of a decade. It’s the benchmark by which mortgage retention is measured.
What’s less consistent is what top performers actually hit. Industry studies aggregate across all LOs, including thousands who don’t actively run a retention motion. Top performers running disciplined recapture systems hit much higher rates, often 2-3x the median. This piece breaks down what the recapture rate actually measures, where the median sits, where the top of the distribution sits, and the four practices that separate the two.
What recapture rate measures
Recapture rate is calculated as: (number of past clients who used the same LO for their next mortgage transaction) ÷ (number of past clients who had a mortgage transaction in the period). The denominator includes anyone in the LO’s funded-loan database who had a credit pull, refinance, or new purchase during the measurement window. The numerator includes only those who closed with the original LO.
The metric matters because past-client recapture is the highest-margin segment of an LO’s pipeline. The lead acquisition cost is effectively zero. The borrower already trusts the LO. The documentation is partially in the system. Closed-loan profitability per past-client recapture is typically 3-5x higher than per cold-acquisition lead.
Where the median sits
The 18% number is striking when you consider the unit economics. An LO with a 500-client database who experiences a typical 20% annual transaction rate among past clients (refis, purchases, HELOCs combined) sees ~100 transactions among that database per year. At 18% recapture, the LO captures 18 of them. The other 82 close with someone else.
If those 82 transactions had stayed with the LO, at an average $5K commission, that’s $410,000 in lost revenue per year per 500 contacts. The bigger the database, the bigger the leak.
Where the top of the distribution sits
Top performers in disciplined recapture systems hit substantially higher rates. The numbers below come from BNTouch’s user-base benchmarks across LOs running the platform’s full retention motion (Equity Alerts + Credit Pull Alerts + Annual Mortgage Review automation + automated nurture sequences).
| LO segment | Recapture rate | Why |
|---|---|---|
| Median LO (no system) | ~18% | Industry baseline. No surfacing of recapture opportunities, ad-hoc outreach. |
| BNTouch user, partial motion | ~25-32% | Has Credit Pull Alerts firing but no Annual Review motion or no Equity Alerts. |
| BNTouch user, full motion | ~35-45% | Equity Alerts + Credit Pull Alerts + Annual Reviews + automated follow-up sequences all running. |
| Top quartile, full motion + scale | ~50-65% | Adds dedicated retention LO or admin. Coverage hits 100%, conversion compounds. |
The gap from 18% to 50% is not a function of rate environment, market conditions, or LO charisma. It’s a function of whether the system is in place. The motions that produce the gap are repeatable, automatable, and don’t depend on the LO being a sales superstar.
The four practices that separate top performers
Practice 1: Coverage discipline
Top performers contact 100% of past clients in the year following close, on the anniversary, with a structured Annual Mortgage Review. Median LOs cover 30-50% of past clients, usually only the ones who happen to come to mind. Coverage is the largest variable. Without 100% coverage, recapture rate caps at whatever portion of the database actually got reached.
Practice 2: Active equity monitoring
Top performers know which past clients have crossed equity thresholds (HELOC eligibility, cash-out, PMI removal) before the borrower realizes it themselves. Median LOs find out only after the borrower goes shopping elsewhere. Equity Alerts surfaces these opportunities ~30 days ahead of when they typically show up in credit-pull data.
Practice 3: Credit pull intercept
Top performers receive an alert within 24-48 hours when a competitor pulls credit on a past client. They call within the first 24 hours, which is when 70% of credit-shopping borrowers ultimately commit to whichever lender called first. Median LOs find out 30+ days later when the loan is gone.
Practice 4: Automated follow-up sequences
Top performers have pre-built drip campaigns for each retention scenario: post-Annual-Review interest, equity-eligible without immediate action, credit-pull intercept, market update for refi-eligible borrowers. Median LOs send one-off emails when they have time, which is rarely.
None of these four practices are individually difficult. The difficulty is running all four consistently across a database that grows every month. Without an automated system, the LO ends up doing 60% of one practice, 40% of another, and 0% of the other two. With an automated system, all four run in the background and the LO’s actual time goes into the high-leverage moment: the call when the alert fires.
The math at top performer levels
For an LO with 500 past clients, moving from 18% to 40% recapture means:
- ~22 additional past-client transactions captured per year (40% of ~100 transactions vs. 18%)
- At an average $5K commission, ~$110,000 in additional annual revenue from the same database
- Compounding effect: each additional past-client retained also refers ~1.4 additional new clients on average, vs. ~0.6 from a non-retained past client
The compounding referral effect is what most LOs miss when they look at recapture rate as a vanity metric. A retained past client doesn’t just produce one transaction. They produce a recommendation network that produces 2-3 additional transactions over the next 5 years.
Why this matters for 2026 specifically
The 2024-2025 rate environment compressed origination volume across the industry. LOs who captured every available transaction stayed profitable; LOs who lost past clients to competitors did not. Recapture rate is the leading indicator of whether an LO is set up for the 2026-2027 origination cycle, when rates ease and refinance volume returns.
An LO running 18% recapture going into a refi cycle leaves the majority of their database’s refi opportunity to other lenders. An LO running 40%+ captures the disproportionate share. The systems that produce 40%+ are the ones that need to be in place before the cycle hits, not after.
If you’re benchmarking your own performance: pull last 12 months of past clients who had any credit pull or transaction. Match against your closed-loan list. Divide. The number you get is your current recapture rate. If it’s below 30%, the system isn’t doing the work it could be.
Common questions
Where is the 18% industry median number sourced from?
Multiple cross-industry sources converge on roughly 18-22% as the median: Stratmor Group’s mortgage industry studies, Optimal Blue’s lender analytics, and Total Expert’s benchmark reports. The exact number varies year-to-year based on rate environment, but the median has been remarkably stable in this range for the better part of a decade.
How do I calculate my own recapture rate?
Numerator: count of past clients who closed a new transaction with you in the measurement period. Denominator: count of past clients who had any mortgage transaction (credit pull, refi, purchase, HELOC) anywhere in the same period. The denominator is the hard part — most LOs don’t have visibility into transactions that happened with other lenders. Credit pull data is the closest proxy.
What’s the difference between recapture rate and retention rate?
Often used interchangeably but they measure slightly different things. Retention rate is typically broader: any continued relationship (newsletter subscribers, opens, follow-up engagement). Recapture rate is narrower: specifically captured the next mortgage transaction. Recapture is the harder, more revenue-relevant metric.
Does recapture rate vary by loan type?
Yes. Conventional purchase recapture is typically lower than refi recapture (borrowers shop more aggressively on purchases). HELOC recapture tends to be higher because borrowers often go to whoever has the easiest application path. Investor (DSCR) recapture is lowest because investors are more rate-sensitive and shop the broker-wholesale channel.
How long does it take to move recapture rate?
Coverage gains show up in the first 90 days (more past clients reached = more captured transactions). Conversion rate gains take 6-12 months as the system matures and the LO refines the conversation script. Database-level effects (referrals from retained clients) take 18-24 months to fully compound.
Want to know your actual recapture rate?
BNTouch can run a recapture audit on your existing database in under an hour. Free demo includes the audit + actionable next steps.
