rates just dropped, which past clients do I call first

Rates dipped this week, and somewhere in your Follow Up Boss and your loan origination system is a short list of past clients who just crossed back into refinance, HELOC, or PMI-removal range. The problem is nobody has that list. By the time someone exports a report and starts calling, the window has usually ticked back up.

Quick answer: When rates dropped, work your database by building a current list of exactly who just became eligible before you call anyone. Pull three data points per past loan (locked rate, loan amount, close date), compare each against today's market, and call the borrowers whose rate sits well above today's first. Skip the full-database blast: a client locked at 5.75% is not a candidate and the touch is wasted.

What actually happened to rates this week

The 30-year fixed fell about 0.20 points week-over-week to roughly 7.26% on October 19, 2026, per the Mortgage Research Center as reported by Forbes Advisor. Bankrate's national 30-year average slipped below 7.5% to 7.47% the week of October 20. That is the kind of quick dip that briefly pulls a slice of a past-client book back into range.

This post does not predict where rates go next, and you should not either when you call. A forecaster cited by Mortgage Daily pegged the 30-year at 7.28% for the week of October 19-23 with a low-confidence falling outlook and a roughly 5.09% 10-year Treasury yield limiting how much further it can drop. Translation for your week: the window is open now and may not stay open. The move matters to you as a trigger, not a rate story.

Who do I call first when rates drop

Call the past clients whose locked rate sits well above today's market, starting with the highest rates and the largest loan amounts, because those borrowers have the most to save and the shortest break-even. A borrower at 7.375% becomes a live candidate the moment rates drop meaningfully below that, while one at 5.75% is not a candidate at all, so marketing to them wastes a touch (ezLoanDocs). Rank your list by the gap between their rate and today's, and the size of the loan, and you have your call order.

The reason first contact has to go out fast is not urgency for its own sake. The same rate drop is visible to every other lender, servicer, and AI digital lender that holds that borrower's information. HousingWire reported that servicers closed 2025 with their best refinance retention since 2014 and still lost two of every three refinancing borrowers, and that for recent loan vintages, AI-powered lenders sit on the payoff request and credit inquiry and recapture the loan before a retail loan officer knows the opportunity exists. The borrower in your database is winnable, but only if someone reaches out before the trigger shows up on someone else's screen.

One more filter before you dial: break-even. Kiplinger and Neighbors Bank both put the point where refinancing becomes worthwhile for most homeowners around a 0.75-point rate drop, where break-even lands under three years. If a past client is only 0.2 points above today, the math may not move them yet. Flag them, but spend your first calls where the gap is real.

The three data points you need, and where they hide

To flag which past clients are candidates, you need three things per closed loan: the locked rate, the loan amount, and the close date, compared against today's market each time it moves (ezLoanDocs). Sounds simple. The reason you do not already have the list is that those three numbers rarely live in one place.

The rate and the close date usually sit in your loan origination system (LOS), the software that documented and closed the loan. The borrower's contact record and tags live in Follow Up Boss or whatever CRM your team runs. A spreadsheet somewhere holds the funded book, or half of it. None of these sync, so the "who's in the money today" query is a manual export-and-cross-reference job, and nobody has time to do it the morning rates dip. That is the operational gap the guides you'll find when you search for this skip: they assume the data already sits clean in one mortgage CRM. For most teams it does not. We wrote more on that scatter in pipeline visibility across four systems.

Segment past clients into the three plays

Once you can see rate, loan amount, and close date together, sort candidates into the three outreach reasons so the message matches the situation. A rate-drop blast fails because a borrower at 7.5% does not need to hear about 6.875% unless it is their 6.875% (loanofficer.ai).

How to build the list before the window closes

Build the current list by pre-tagging your past clients once, by loan type, rate, close date, and program, so a segmented list can be deployed the moment rates move instead of assembled from scratch (Scotsman Guide). The scramble loses because you start the export after the dip; the teams that win started the tagging months ago.

That last point is where most reactivation dies. The list gets built, a few calls go out, and the follow-up stops the moment a borrower says "let me think about it." The contact rate problem operators keep flagging (stuck around 11%, lifted to 28% by one 45-LO team purely through tighter multi-channel follow-up over 7-plus touches) is a follow-up problem, not a list problem.

Manual scramble vs. a system that watches the trigger

The honest comparison is between doing this by hand when you remember to, and having something watch for the trigger and hand you the ranked list the morning it happens.

This is the gap SalesOS is built for. SalesOS, the part of our AI Cofounder we implement for sales-led teams, keeps your CRM and sales data in sync and can surface and work the stalled past-client list in your team's own voice when a trigger like a rate dip hits, instead of waiting on a weekly meeting. It follows the texting and outreach rules your team sets; your compliance people decide what goes out automatically and what waits for a person. You can see how AI Cofounder works if you want the full picture of how it runs inside the Claude or ChatGPT account you already pay for.

Common pitfalls when working the database on a rate move

The fastest way to burn a rate dip is to blast the whole book. A client locked below today's rate does not want to hear from you about refinancing, and the touch costs you credibility for the next real opportunity. Match every alert to the borrower's actual rate.

For a longer walkthrough of working a large candidate list without hiring, see working 50+ refi candidates when rates drop.

FAQ

How many past clients actually become candidates on a small rate dip?

On a 0.2-point dip, only the slice of your book locked well above today's market crosses into real refinance range, usually the loans from your highest-rate vintages. UWM's Alex Bressler told National Mortgage Professional that a rate drop can flag 50-plus past clients, and one loan officer has no shot at calling them all, which is exactly why the ranked list and the follow-up matter more than the alert.

Can this run off the Follow Up Boss and LOS I already pay for?

Yes. The point is to keep the CRM and sales data you already run in sync and surface the candidate list from them, not to move you onto a new mortgage CRM. SalesOS works with the tools your team already has rather than replacing them.

Do I need to predict where rates go to work my database?

No. You work the database off the move that already happened, not a forecast. Build the current in-the-money list, call the biggest gaps first, and review each borrower's situation; leave the rate prediction out of the conversation entirely.

How do I avoid breaking texting rules when I reach out fast?

Any automated outreach follows the rules your team and your compliance people set, including the texting restrictions your CRM already enforces. Decide in advance what goes out automatically and what waits for a person to send, and let your compliance team make the call, not the tool.

If your past-client list is scattered across your CRM, your LOS, and a spreadsheet, and the next rate dip is going to find you exporting reports again, that is the exact gap worth a conversation. Book a demo and we'll walk through how your team could have the ranked list ready the morning rates move.