Mortgage Renewal Follow-Up: An Operating Guide for Brokerages
Most brokerages do not have a renewal problem. They have a process problem that shows up at renewal.
A brokerage with five years of funded deals behind it is carrying a renewal book whether or not anyone is managing one. The files exist. The maturity dates exist. What usually does not exist is a repeatable answer to a specific question, asked every week:
Which borrowers should someone contact about their renewal, and who is doing it?
When that question has no owner and no routine, renewals do not get lost dramatically. They get lost quietly, one at a time, to whichever lender sent a renewal letter first.
This guide is about the operating process rather than the borrower conversation. It assumes you already know how to advise someone at renewal. The harder problem is making sure the conversation happens at all.
What renewal follow-up actually means for a brokerage
For a borrower, a mortgage renewal is an event. For a brokerage, it is a recurring operational obligation against a growing list.
That distinction matters because the two things fail differently. A borrower fails at renewal by signing the first offer they receive. A brokerage fails at renewal by never surfacing that borrower's maturity date to a human in time to do anything about it.
So the work splits into four jobs that are easy to conflate:
- Knowing which renewals are approaching, and when.
- Deciding which of them deserve attention first.
- Doing the outreach, and recording that it happened.
- Noticing when any of the above stops working.
Most brokerages are reasonably good at (3) once someone is actually looking at a name. The revenue leaks out of (1), (2) and (4).
Why renewals get missed
Renewals rarely go missing because someone decided not to bother. They go missing because of ordinary operational entropy:
- The maturity date is stored, but not surfaced. It sits in a column nobody sorts by. Data you cannot see on a Monday morning is not operational data.
- Renewals compete with live deals, and lose. A file in underwriting has a lender waiting. A renewal 90 days out has nobody waiting. Urgency wins, every week, until the renewal is 10 days out and now urgent for the wrong reason.
- Nobody owns the list. Where renewal follow-up is "everyone's job when they have a minute", it is nobody's job on a busy week — and busy weeks are most weeks.
- The record of contact is inconsistent. If two people cannot tell whether a borrower was already called, one of two things happens: the borrower gets contacted twice, or not at all.
- The book grew. A process that worked by memory at 80 past clients does not work at 400. Nothing broke; the method simply ran out of headroom.
None of these are failures of effort. They are failures of a system that was never designed, only inherited.
Where traditional tracking breaks down
Most brokerages already have somewhere the data lives — a CRM, a lender portal, a spreadsheet, or some combination. The common failure is not absence of data. It is that storage and prioritisation are different jobs, and most setups only do the first.
A system that stores a maturity date answers "when does this mature?" once you already know which borrower you are asking about. A renewal process needs the inverse: given everything in the book, who should we work this week? That is a sorting-and-ranking question, and a record store does not answer it by default.
The practical symptom is easy to recognise. If producing a list of this week's renewal priorities requires someone to export, filter, sort and eyeball a spreadsheet for twenty minutes, then that list gets produced when someone has twenty minutes. Which is to say: not weekly.
What information a brokerage actually needs
You need less than you might expect. A workable renewal process runs on a handful of fields per borrower:
| Field | Why it matters |
|---|---|
| Borrower name and contact | Obvious, but it must be current — a stale mobile number is a missed renewal |
| Maturity or renewal date | The anchor for every timing decision |
| Original close date and term | Lets you derive the renewal date when it was never recorded directly |
| Loan amount | The main input into prioritisation |
| Last contact date | Prevents both silence and double-contact |
| Consent status | Determines whether you may contact them electronically at all |
| Owner | Who is responsible for this specific borrower |
| Outcome | What happened, so next week's list is different from this week's |
If any of these is missing across most of the book, fix that before designing a cadence. A follow-up process built on data you do not trust produces a list nobody acts on, which is worse than no list — it burns the team's confidence in the process itself.
How to segment upcoming renewals
Not every renewal deserves the same treatment, and pretending otherwise is what makes renewal outreach feel like a mass-email chore.
Three segmentation axes do most of the useful work:
By time to maturity. This determines when, and is covered in detail in the renewal cadence guide.
By value. Loan size is a blunt instrument but an honest one. A $780,000 mortgage and a $190,000 mortgage do not warrant identical effort, and pretending they do means the larger one gets the same automated email as the smaller one.
By relationship state. A borrower you spoke to four months ago is a different conversation from one who has heard nothing since funding. A borrower who came through a referral partner is a different conversation again — that renewal has a relationship attached to it.
A brokerage does not need a scoring model. It needs to stop treating a list sorted by date as if it were sorted by importance.
The operating rhythm
The widely used cadence in the Canadian market works backwards from maturity in roughly 30-day steps: an early contact around four months out, then progressively more direct conversations as the date approaches.
Industry guidance broadly supports engaging early rather than late. Canadian Mortgage Professional has described brokers engaging clients 120–150 days out from maturity, with some doing a soft touch around six months and getting serious near the four-month mark.
Treat that as an operating convention, not a rule. It is not a regulatory requirement, and it is not the same thing as the notice period a lender owes a borrower — those are separate obligations that vary by jurisdiction and product. What the cadence gives you is a repeatable structure so that "when should we contact this person?" stops being a judgement call made from scratch every time.
The detail of each touch is covered in the 120/90/60/30 breakdown.
How to prioritise
Within any given week you will have more renewals in range than you have hours. Prioritisation is the whole game.
A workable ordering, in decreasing weight:
- Time pressure. A renewal 30 days out beats one 110 days out, regardless of size.
- Value at stake. Among borrowers in the same window, larger balances first.
- Silence. A borrower with no contact recorded since funding is a higher risk than one you spoke to last month.
- Relationship leverage. A referral partner's client carries a second relationship.
The specific weighting matters less than having one and applying it consistently. An imperfect ranking applied every week beats a perfect ranking applied when someone gets to it.
Who should own follow-up
Renewal follow-up fails as a shared responsibility. It works when a named person owns the list.
That does not mean one person makes every call. It means one person is accountable for the list being produced, worked and reviewed. In a small brokerage that is often the principal. In a larger one it is usually an operations or client-care role — and this is frequently a better fit, because the work is administrative discipline punctuated by advice, not the reverse.
The distinction worth protecting: producing the list and working the list are different jobs. If the person making the calls is also the person assembling the list, the list gets assembled only when calls feel worth making.
What to automate and what not to
A useful dividing line: automate the identification, keep the conversation human.
Reasonable to automate:
- Deriving renewal dates from close date and term
- Surfacing who enters each window this week
- Ranking that list
- Recording that an outreach happened
- Early, low-stakes touches at the outer edge of the cadence
- Flagging borrowers who fell out of the process
Keep human:
- The 60- and 30-day conversations, especially on larger balances
- Anything where the borrower's circumstances have changed
- Anything involving a referral partner relationship
- Every negotiation
Automating the wrong half is a recognisable failure. A brokerage that automates the conversation sends four templated emails to a borrower with an $800,000 mortgage and loses them to a lender who phoned. A brokerage that automates identification and calls the top of the list keeps them.
What to measure
Four numbers, reviewed monthly, tell you whether the process is real:
- Coverage — of borrowers who entered a renewal window this month, what share received the intended contact? This is the number that exposes a broken process fastest.
- Contact-to-conversation rate — how many touches produced an actual two-way conversation.
- Retention at maturity — of renewals that matured this month, how many stayed with the brokerage.
- Time to first contact — days between entering the outer window and the first recorded touch.
Coverage is the one to start with. It is the only one that isolates process from market conditions. A brokerage can lose renewals in a competitive rate environment while running a perfect process; it cannot lose them to 40% coverage and blame the market.
Common failure modes
- The list exists but is never worked. Usually an ownership problem, not a data problem.
- Everyone gets the same email. Efficient, and it trains borrowers to ignore you.
- Contact is not recorded. Within two weeks nobody can tell what has been done.
- Only the next 30 days are visible. By 30 days out, the borrower has often already had a letter from their lender.
- The process runs on one person's memory. It works, until that person is on holiday, or leaves.
- Consent is assumed. An empty consent field is not permission. Absence of evidence is not consent.
What a Weekly Renewal Action List changes
The single highest-leverage change most brokerages can make is to stop treating renewal follow-up as a project and start treating it as a standing weekly output.
One list. Produced the same day each week. Ranked. Owned. Reviewed.
Consider a hypothetical brokerage with 400 funded borrowers and 65 renewals maturing in the next six months. Approached as a project, that is a daunting, easily deferred pile. Approached as a weekly list, it is perhaps eight to twelve names a week — an hour of work, with the highest-value ones at the top.
The pile does not change. What changes is whether it is workable on a Tuesday. How to build that list covers the fields and the routine.
When a brokerage needs an audit
If you can answer these from memory, your process is probably fine:
- How many borrowers mature in the next 120 days?
- Which five are the largest?
- Which of them has nobody spoken to since funding?
- Who is responsible for contacting them this week?
If those questions require a spreadsheet and half an hour, the renewals are not the problem — the visibility is. And the revenue at stake is not hypothetical: it is already in the book, attached to borrowers you have already earned once.