How to Size the Customer List You're Not Calling
A 20-minute exercise for a lending COO: pull three customer lists by branch, check who was called in 90 days, and see the size of the list nobody is working.
Pull three lists by branch: customers who paid off in the last 24 months, current borrowers in good standing, and applications that never closed. Then check which of those people anyone at your company actually called in the last 90 days. Everyone left over is your idle list. Count it branch by branch. That gap is the opportunity, measured with your own numbers.
Pull three lists from your loan system, by branch. Customers who paid off in the last 24 months. Current borrowers in good standing. Applications that never closed.
Then ask one question about every name: did anyone call this person in the last 90 days?
Everyone left over is your idle list. Count it branch by branch. That's the customer list you're not calling, and it's usually bigger than anyone in the building expects.
You can do this in about 20 minutes. Here's how, with a worksheet you can copy.
Why nobody in the building knows this number
Ask most lending COOs how many past customers they have. You'll get a number, roughly.
Now ask how many of them were called last quarter. The room gets quiet.
It's not that anyone is hiding it. It's that the two numbers live in different places. The customer count lives in the loan system. The calling, if it happened, lives in a branch employee's memory, a sticky note, or a call log that nobody exports.
Picture the Monday call with your regional managers. Each one reports on applications, fundings, delinquencies. Those have owners and reports. Nobody reports on "past customers we reached this week," because there's no report for it.
So the list doesn't look like a problem. It looks like nothing at all. That's why past customers are the cheapest loans you'll make this year, and also why they're the easiest to ignore.
Step 1: Pull three lists by branch
Ask whoever runs reports on your loan system for three pulls. Keep them separate. Keep them by branch.
Customers who paid off in the last 24 months. They were underwritten, they paid, and they know your branch. Two years is a reasonable window. Older than that and phone numbers start going stale. If your team can split this into "last 6 months" and "6 to 24 months," even better. The recent ones are warmer.
Current borrowers in good standing. People who are paying on time right now. Whether and when you can offer them anything depends on your products, your policies and your state. This list is here to show you how many of them your branches actually talk to, outside of payment reminders.
Applications that never closed. Someone asked you for a loan and it didn't happen. Maybe they stopped answering. Maybe a document never came in. Maybe the branch was busy the day it arrived. Some of these were declined for good reasons, so filter those out according to your own rules.
For each list you want two things per branch: how many people are on it, and how many have a usable phone number. The second number is the real one. A name without a working number isn't a customer you can call.
Step 2: Ask who called each of these people in the last 90 days
This is the step that usually hurts.
For every person on those three lists, you want to know whether someone at your company placed a call to them in the last 90 days. Not a mailer. Not an email. A call.
If your calls are logged in a CRM or phone system, match on phone number or customer ID and count.
If they're not logged anywhere, you have your answer already. It's not "we don't know." It's "zero that we can prove." And that's a finding worth writing down, because it means nobody could manage this even if they wanted to.
A few rules keep the count honest:
- A call means a logged attempt. "I think someone called a few of those" doesn't count.
- Payment reminders and collection calls don't count for the paid-off and application lists. They're a different conversation.
- Ninety days, not "recently." Pick the window and stick to it.
Subtract the called from the total. What's left is the idle list.
The worksheet
Copy this into a spreadsheet. One row per branch. Fill it with your own numbers.
| Branch | Paid off, last 24 mo | Good standing, current | Applications never closed | Total with a usable phone | Called in last 90 days | Idle (not called) |
|---|---|---|---|---|---|---|
| Branch A | — | — | — | — | — | — |
| Branch B | — | — | — | — | — | — |
| Branch C | — | — | — | — | — | — |
| All branches | — | — | — | — | — | — |
Two columns matter most. Total with a usable phone tells you how many people you could reach. Idle tells you how many nobody is reaching.
When you look at it by branch, patterns show up fast. One branch has a huge paid-off list and almost no calls. Another is calling, but only its newest applications. Those are management conversations you couldn't have before, and they're the start of the kind of branch-level reporting a COO should see weekly.
How to estimate what the gap is worth, with your numbers
You're probably wondering what the idle number turns into. Here's the honest answer: nobody knows until the list gets called. Anyone who gives you a precise forecast before that is guessing.
What you can do is set up the math so it runs on your numbers, not someone else's:
- Idle list with a usable phone. From the worksheet.
- Calls it takes to work that list. Most people won't answer the first time. Plan on calling no-answers again, so calls will be higher than the number of people.
- Live conversations that turn into a real interest in a loan. Only calling tells you this.
- Your own average loan and your own funding rate on customers who come back. Your finance team knows these. We don't.
There is one public reference point we can share. It's what happened at Sunset Finance, not a prediction for you. In their pilot, calling past customers who had already borrowed once, with a pre-approved offer of up to $1,000, produced about 8 live transfers for every 1,000 calls placed. You can see the full picture in the Sunset Finance case study.
So if you want a rough illustration, take the number of calls your idle list would need, divide by 1,000, and multiply by 8. Then treat the result as a question to test, not a number to budget.
Why the gap gets bigger every month
The idle list isn't standing still while you think about it.
Phone numbers go stale. People move and change carriers. The 24-month-old paid-off customer is harder to reach than the 2-month-old one.
Other lenders keep marketing. The Federal Trade Commission explains that creditors can set requirements based on credit report information, like borrowing and payment history, and ask a credit bureau for a list of people who meet them. A borrower who just paid you off on time can land on someone else's list.
Applications go cold fastest. Harvard Business Review's study of 1.25 million sales leads found firms that tried to reach a lead within an hour were nearly seven times as likely to qualify it, meaning a meaningful conversation with the decision maker, as firms that tried an hour later. That research was on web leads across industries, but every loan officer knows the feeling. The same thing happens to the applications that arrive when nobody can answer, which we cover in what happens to loan applications that arrive after 5 p.m.
What to do once you have the number
Once the idle column is filled in, the question stops being "is there an opportunity?" It becomes "who is going to call these people?"
Usually the answer can't be the branch. The staff are already covering the counter, payments, collections and the phones. That's the gap an AI voice agent fills well. It calls the whole list during branch hours, from a local number, calls no-answers again, and transfers the interested customers live to their own branch while someone is there to pick up. Every outcome goes back to your CRM, so next quarter the "called in last 90 days" column fills itself in. How to reactivate paid-off borrowers without a mail campaign walks through that step by step.
Before any list gets called, check your consent and do-not-call rules with your compliance team.
Then start with one branch, not the network. Every month you wait, the list you just pulled gets a little older. If you want to hear what your customers would hear, the AI will call you. Leave your number.
Questions lenders ask
Which customers should go on the idle list?
Three groups, by branch: customers who paid off in the last 24 months, current borrowers in good standing, and applications that never closed. Remove anyone who was actually called in the last 90 days.
What counts as being called?
A logged call attempt to that person in the last 90 days. A mailer doesn't count, and neither does a call someone remembers making but nobody recorded.
How long does this take?
About 20 minutes if someone can pull reports from your loan system by branch. Longer if call activity isn't logged anywhere, which is itself a finding.
How many live transfers would our list produce?
Nobody can tell you that honestly before calling it. At Sunset Finance, the pilot produced about 8 live transfers for every 1,000 calls placed. That is what happened there, not a prediction for your branches.