The fifth inquiry

One of the four lenders the couple approached ran the credit application through an intake system that, for this particular product line, defaults new files to a generic instalment-loan purpose code unless a broker manually overrides it. Nobody caught the override that day, so that single application landed on the bureau file coded as an instalment-loan enquiry rather than a mortgage enquiry — and an instalment-loan enquiry does not fall inside a mortgage rate-shopping window.

None of this changed the couple's actual creditworthiness. It changed what an automated system saw: a current lender's own rate card prices this insured product by credit-score band, and the file's automated score sat at 609 — inside the lender's lower pricing tier, not because anything was genuinely wrong with the file, but because a data-entry default at a different lender's intake desk, three weeks earlier, had been read as a fifth, unrelated credit-seeking event.

The purchase itself is a standard insured file at 10% down; the pricing tier, not the ratios, was what the miscoded inquiry put at risk.

This lender's own rate card prices by credit-score band: the lower band (under 620, where the miscoded score of 609 landed) carried a 4.84% contract rate; the band the file actually belonged in (620–679) carried 4.64%.

Both bands clear CMHC's 39%/44% caps comfortably — the ratios were never in question. The $46-a-month gap between the two payments, for the life of the term, was the entire cost of one lender's intake default, not of anything about the couple's actual credit behaviour.

An FSRA-licensed Ontario mortgage agent pulled the full bureau file, not just the score, once the automated number came back lower than the file's history suggested it should.

First, read every inquiry's purpose code and date individually, rather than accepting the inquiry count at face value — this surfaced four inquiries dated within days of each other under a real-estate financing code, and a fifth, from the same week, coded as an instalment loan.

Second, contacted that lender's credit department directly and obtained written confirmation that the application had, in fact, been a mortgage pre-qualification that its own intake system defaulted to the wrong purpose code.

Third, presented the marked-up bureau file and the lender's own confirmation letter to the current lender's underwriter, who manually recognized the fifth inquiry as part of the same shopping event rather than requiring a formal, weeks-long bureau dispute to resolve it.

The underwriter accepted the correction and priced the file in the 620–679 band at 4.64%, not the lower band's 4.84%. GDS at the qualifying rate came in at 35.1%, comfortably inside CMHC's 39% ceiling.

The exact point-value impact of any single inquiry varies by scoring model and file depth; the figures here illustrate the shape of the problem — a miscoded inquiry outside the rate-shopping window — not a universal per-inquiry score deduction.

• 01A rate-shopping window only protects inquiries coded correctly. Canadian bureaus bundle multiple mortgage inquiries within roughly four to six weeks — but only if each one is actually coded as mortgage financing.
• 02Read the bureau file's inquiry purpose codes, not just the count. Four properly-coded inquiries and one miscoded one look identical in a simple tally of 'five inquiries.'
• 03An automated score can be technically accurate and still be based on a data-entry error elsewhere. The fix here was a corrected classification, not a bureau dispute or a credit-repair process.
• 04Inquiries carry more weight on a thinner file. The same miscoding would likely have mattered less on a borrower with a decade of bureau history behind them.