Treadstone Associates
Case File № 614 · Bruised Credit & Consolidation

128 points that were never real

a Collingwood file nearly walked away from its own approval

A Collingwood buyer's free credit-monitoring app showed a comfortable 742 for months. The mortgage lender's own bureau pull came back at 614 -- a different, mortgage-specific scoring model, not an error -- and the client nearly delayed closing trying to fix a 128-point gap that had nothing to do with the file the lender actually underwrote.

OntarioInsured · PurchaseFiled August 9, 20265 min read
742

the score shown by the client's own free credit-monitoring app for months

614

the mortgage-specific score the lender's own bureau pull actually returned, the same week

128 pts

the gap between them -- a different scoring model, not a real decline

Anonymized illustration. The borrowers, dollar figures, and rates in this file are an illustrative composite — no real client is identifiable, and any rate shown is illustrative, not a quote. The rules are real: every regulatory figure is cited to its source in the Sources section, and the math computes exactly as shown.

№ 01

The client

A buyer in Collingwood purchased a $358,000 home at 5% down, insured, having tracked their own credit for months through a free monitoring app.

Purchase price

$358,000, Collingwood

5% down, insured

App score

742

Shown by a free consumer credit-monitoring app

Lender's own score

614

The mortgage-specific model the lender actually pulled

Combined income

$7,700/month

№ 02

The problem

Free consumer credit-monitoring apps license a scoring model built for consumer education, not mortgage underwriting. A mortgage lender's own bureau pull uses a different, mortgage-specific model -- built from the same underlying file, but weighted differently -- and the two numbers are not the same product.

What a 128-point gap actually meant

  • The app's 742 was a real number from a real scoring model -- just not the one any lender was going to rely on
  • The lender's own bureau pull returned 614, still comfortably above CMHC's 600-score floor for an insured file
  • Neither number was wrong. They were reading the same underlying credit file through two different, unrelated models

The client read the 128-point drop as evidence something had gone wrong in the weeks since they last checked. Nothing had.

№ 03

The numbers

At the lender's own score -- the only one that actually governed this file -- the ratios were never close to a problem.

The insured purchase, on the score that matteredAmount
Base mortgage (95% of purchase price)$340,100
CMHC premium (4.00% at 95% LTV)+$13,604
Total insured mortgage$353,704
Ratio check at the qualifying rateFigure
Payment at the qualifying rate (7.35%), 25 years$2,554/mo
GDS (payment + $265 tax + $105 heat) ÷ $7,700 income38.0%
TDS (GDS numerator + $200 car loan) ÷ $7,700 income40.6%

38.0% and 40.6% sit comfortably inside CMHC's 39% GDS and 44% TDS maximums, in the range Canadian mortgage credit-score data shows clears without difficulty. At 614, this file was never in danger -- the only real risk was the client walking away over a number that was never in play.

№ 04

The solution

A mortgage agent licensed under Ontario's Mortgage Brokerages, Lenders and Administrators Act separated the two scoring products in writing before the client spent more time chasing what looked, on the surface, like a disputed credit item.

First, confirmed which scoring model the app licensed and which model the lender's bureau pull actually used. The two are built by different processes for different audiences, and a gap between them is normal, not diagnostic.

Second, confirmed the lender's 614 score in writing, well above CMHC's 600 floor, and walked the client through exactly why no dispute, paydown, or waiting period could -- or needed to -- close a gap that was never a data error.

Third, kept the file moving on its original timeline rather than letting a misunderstood number introduce a delay the file never actually needed.

Written confirmation of the lender's own bureau-pulled score
A plain-language explanation of the app-score vs. lender-score gap, for the client's file
Standard insured-purchase documentation for income, down payment and credit
No credit-repair or dispute action taken, since none was needed
Closing kept on its original schedule
№ 05

The outcome

The purchase funded insured at 38.0% GDS and 40.6% TDS on schedule, once the client stopped chasing a 128-point gap that had nothing to do with the file the lender actually underwrote.

Both ratios sit comfortably inside CMHC's 39% GDS and 44% TDS maximums; the 614 score itself was never close to CMHC's 600 floor.

№ 06

What to take from this file

  • 01A free consumer app's credit score and a lender's own mortgage-specific bureau pull are different products. A gap between them is a model difference, not a data error to dispute.
  • 02Confirm the lender's own score early, in writing. It is the only number that actually governs an approval -- not whatever a client's banking or monitoring app happens to show.
  • 03A score gap this size can trigger real client anxiety even when nothing is wrong. A short, plain explanation up front prevents a self-inflicted delay.
  • 04Do not let a client 'fix' a problem that doesn't exist. Disputing accurate accounts or rushing a paydown to chase a consumer-app number can create real complications for no real benefit.

Sources

Every regulatory figure in this file traces to one of these primary sources. Client details and anything that varies by lender are illustrative, as flagged below.

Illustrative in this file — lender-specific, not rules:

  • 5.35% contract rate — rates move daily; not a quote.
  • the 742 / 614 scores and the 128-point gap — which consumer app and which bureau/scoring model a lender uses both vary; this file's own particular gap is not a universal figure.

Authority & provenance

How this case file was built

We publish the origin, the verification method and the reviewer for every case file, so you can judge how far to trust it before you rely on it with a client.

Where it comes from

Derived from files handled by Treadstone’s fulfillment desk and from scenarios contributed by partner brokerages. Names, employers, exact amounts and dates are changed so no client or file is identifiable.

Provenance: Composite — a pattern seen repeatedly on fulfilled files, not a single transaction.

What is verified

Every regulatory figure traces to a primary source listed above and was checked against it on the date shown. The arithmetic is recomputed by machine on every rebuild.

Anything that varies by lender is labelled illustrative rather than stated as a rule.

Who reviewed it

Reviewed for Canadian regulatory accuracy before publication, and re-checked whenever a cited rule changes.

Reviewed by: Nicholas Parson, Treadstone Associates — reviews every case file before publication.

First published 9 August 2026Rules last verified 9 August 2026Next scheduled review 9 February 2027

This case file is professional reference material for licensed mortgage professionals. It is not advice to a borrower, and it is not a lender commitment. Insurer rules, qualifying rates and provincial taxes change — confirm the current position with the insurer, regulator or lender before you rely on any figure here in a live file.

Treadstone fulfillment

Files like this are daily work for our desk.

Document collection, ratio math under multiple treatments, lender placement notes, and submission-ready packaging — for Canadian mortgage brokers who would rather be in front of clients.