The firm

A lending desk, drawn in section

Chordline is a fictional company built to present a website template for commercial mortgage lending. This page describes how a real desk of this kind is organised — and is careful to claim nothing about this one, because there is no this one.

There is no team. No named individuals, no photographs of people presented as staff, no biographies, no headcount and no founding year appear anywhere on this site. Inventing a person is a different order of fiction from inventing a company.

The idea the whole site is built on

A commercial mortgage is a span. The property’s net operating income is the top chord, in compression. The debt service is the bottom chord, in tension. The depth between them — the coverage — is what lets the structure carry anything at all. A truss with no depth is a flat bar, and a loan with no coverage is a default waiting for a bad quarter.

That is not decoration. It is the reason the header is a chord with panel points on it, the reason cards are built as extruded steel sections rather than tiles with shadows, and the reason the hero instrument draws a truss that gets deeper as the coverage ratio rises. If you move the coverage floor to 1.00× the truss collapses to a line, which is exactly what it means.

What a desk like this actually is

Four functions, deliberately separated. On a small desk one person may cover two of them, but the separation between origination and credit is not negotiable anywhere — the person who wants the deal to happen cannot be the person who approves it.

Origination

Takes the first call, sizes the deal on the spot, and says no early when no is the answer. Owns the relationship from the conversation to the term sheet.

Underwriting

Reconciles the trailing twelve to the rent roll, tests the rollover, adjusts what has to be adjusted, and writes the credit presentation.

Credit

Approves, declines, or approves with conditions. Independent of origination by design — the person who wants the deal is not the person who approves it.

Closing

Third-party reports, legal, title, insurance, escrows and the wire. The stage where a well-prepared file overtakes a better one.

Four things the copy on this site is written around

  1. 01
    Name the binding constraintA sponsor told “we can do 65%” when debt yield is really what caps the deal will spend money on the wrong lever. Saying which test binds costs nothing and changes what someone does next.
  2. 02
    Say no earlyThe expensive no is the one that arrives after the appraisal has been paid for. If a deal is not going to work, the useful moment to find out is stage two, not stage five.
  3. 03
    Publish structure, not ratesA rate nobody has been underwritten for is a guess with a decimal point in it. Coverage floors, debt yield floors, reserve mechanics and prepayment structures are checkable and stable, so those are what this site explains.
  4. 04
    The awkward part is the useful partThe deferred roof, the lease expiring in year three, the partner who wants out. Files that lead with those get read faster than files where they are found.

Where this site is honest by omission

Several things a real lender would publish are missing here, and the omission is deliberate rather than an oversight. There is no NMLS identifier, no state lending licence number and no registration, because a made-up number in a real registry format could collide with a working broker’s. There is no closing-time figure, no approval rate, no volume figure and no review count, because every one of those would be invented. There is no map and no coordinates, because a geocoder handed an invented address drops a pin on somebody’s real building.

The disclosures page lists all of it in one place, including the things that are true.