What is fixed, what we assume, and what still has to be tested

A source fact comes literally from a dated document: a sales brochure, a lease, a valuation or a title extract. We always show which document it is and when it dates from. If the brochure says 'approximately 40 m² gross', we write approximately 40 m² gross, not 40.5.

An assumption is a choice made by the model: interest, indexation, exit yield, owner costs, tax rate, hold period. Assumptions are marked as such and are editable in the calculator. What the source says never changes with them.

Missing evidence is named rather than filled in. If no executed lease has been supplied, we say so — and where a lease date is only known by year, the exact date we model is labelled as an assumption.

The year-one net property yield sets the first year's net rental income against the asset acquisition cost, including tax and costs but before any financing fee. Cash-on-cash does the same for the cash flow after interest and scheduled principal, against the equity put in. Both describe a single year.

The IRR discounts every cash flow on its actual date, including the sale at the end. It is sensitive to the exit yield and the hold period, and therefore less certain than a cash return. We calculate the IRR annually over actual dates; a monthly return is never multiplied by twelve.

Where a cash flow series could mathematically have more than one internal rate of return, we only report an IRR when a formal uniqueness test passes. Otherwise we say that no unique IRR has been established and present the price at the chosen discount rate instead.

If there is no support for the period after the sale date — for example because the lease has ended by then — we show no exit value and no IRR. The operating cash flow stays visible.

Every analysis starts as a draft, with a version number and the status 'not reviewed'. At that stage the figures exist to test the model, not to base a decision on.

Before publication the source facts are checked against the original documents, the assumptions are tested against market information, and it is recorded who carried out the review. Only then does the status change.

Reletting after the end of a running lease is an explicit assumption. Where it is modelled, the vacancy period, the new rent, the term and the one-off costs are all stated, and the exit is blocked whenever the twelve months after a sale are not fully supported.