Database Recapture After the Trigger Lead Ban: How to Mine Your Past Borrowers

Database recapture: how to actually mine the asset you already paid for

Mortgage acquisition is two different problems pretending to be one. Cold acquisition is a marketing problem. Database recapture is an operations problem. Most loan officers solve the first one decently and ignore the second one almost entirely. After the Homebuyers Privacy Protection Act closed the trigger-lead pipe, the recapture side stops being a nice-to-have and becomes the primary growth lever.

This post is the operational deep-dive on the recapture side. The pillar covers the regulatory why. This covers the mechanical how.

Why most LOs ignore their database

Three reasons, in honest order:

1. The work isn’t visible. Cold acquisition has obvious metrics (leads, cost per lead, cost per close). Database work has muddy attribution. A past borrower who closes again could have come from anywhere. Without a CRM that tracks the trigger event (call, alert, email opened, form submitted), the wins look like luck instead of strategy.
2. The work is unglamorous. Calling someone you closed for in 2022 to ask if they want to refi feels less productive than chasing a fresh lead. The opposite is true. Past borrowers convert at 3-5x the rate of cold leads at a fraction of the cost.
3. The work compounds slowly until it doesn’t. Year one of database mining looks like a trickle. Year three looks like a river. Most LOs quit before the compounding shows up in P&L.

The recapture rate math

Recapture rate is the percentage of your past borrowers who use you again when they get a new mortgage. The 2024-2026 data, per STRATMOR Group and various IMB benchmarking surveys, puts the industry average around 18-22%. Top performers are at 40%+. Some IMB shops with disciplined recapture programs hit 50-60%.

For an individual LO with 1,000 past borrowers, here’s the rough math:

• Annual transaction rate of past borrowers (refi + purchase combined): roughly 8-12% in a typical year, swings higher when rates drop.
• 1,000 borrowers × 10% = 100 borrowers transacting per year.
• At industry-average 20% recapture: 20 closed loans per year from your existing database.
• At top-performer 40% recapture: 40 closed loans per year.
• Difference: 20 closed loans × $400K avg loan × 1% commission = roughly $80,000 in additional production from the same database.

This isn’t theoretical. The IMBs running disciplined recapture programs are pulling these numbers right now. The ones who aren’t are leaving the same dollars on the table month after month.

For deeper benchmarking by team size and market, see our 2026 LO recapture rate benchmark.

Segmentation: the inputs that drive everything else

You can’t run one cadence against a flat list. The database has to be segmented by the signals that determine when a borrower is ready to transact again:

• Locked rate vs. current rate. Anyone locked above 6.5% in 2022-2023 is a refi candidate the moment the 30-year drops 50 basis points below their lock. Tag them.
• Years since closing. 2-5 years out is the typical refi window in normal markets. Less than 2 years is too early in most cases. More than 7 years and the relationship has decayed.
• Loan type. Adjustable-rate borrowers approaching reset, FHA borrowers eligible for streamline refi, VA borrowers for IRRRL — each has different timing and pitch.
• Life-event signals. Marriage, divorce, new job, kid heading to college — all correlate with mortgage activity. Public records and social signals can flag these.
• Credit-pull events. The strongest signal. When your past borrower has their credit pulled by another lender, they’re actively shopping right now. Tools like BNTouch’s Credit Check Alerts surface this within 1-2 days of the event.

Contact cadence by segment

Different segments get different cadences. A high-intent credit-pull alert demands a phone call within an hour. A 2022-locked-at-7% borrower in a stable rate environment gets a quarterly check-in. The cadence has to match the signal strength.

Automated triggers vs. manual outreach

The recapture program has to be a mix. Some triggers are objective (rate drops below threshold, ARM hits reset month, anniversary date) and should fire automated email/SMS sequences. Others (credit-pull alert, life-event signal) are human moments that need an LO on the phone.

The split that works for most teams: automation handles the wide-coverage maintenance contact (quarterly rate updates, anniversary touches, drip nurtures), and the LO handles the high-signal events. Trying to make automation do everything produces high open rates and zero conversions. Trying to make the LO do everything produces 4% database utilization and missed deals.

For the playbook on annual mortgage reviews specifically (the highest-leverage automated touch in most programs), see our annual mortgage review guide.

The compounding effect

Year one of disciplined database mining produces an extra 5-10 closings on top of normal flow. Year two doubles. Year three, the database itself has grown (more closed loans = more borrowers in the recapture pool), and the cadence is dialed in. The LOs running this program for five years compound to a place where 50%+ of their annual production comes from past borrowers, not new acquisition.

That’s the moat HBPPA just made strategic. Lenders without database discipline are now the ones paying $80-300 per opt-in cold lead. Lenders with disciplined recapture are paying $10 per high-intent past-borrower lead via credit-pull monitoring. Same close rate, dramatically different unit economics.

For the right CRM features that make this discipline practical, our 2026 buyer’s guide covers which platforms support real recapture workflow vs. which ones treat the database as a static contact list.

Common questions

See how Credit Check Alerts work →
Book a Demo →