Documenting Comparables in a Valuation Report

A valuation is only as strong as the trail it leaves behind. The figure in the conclusion may be correct to the krone — but if no one can follow the calculation backwards, to the sales that support it, and on to why exactly those sales were chosen and how they were adjusted, the report will not survive its first critical review. It is rarely the valuation itself that sinks a case before a credit committee, an auditor during a financial audit, or a counterparty in a dispute. It is the gaps in the documentation.

That is why documenting the basis of comparison in a valuation is not a formality you add at the end. It is the discipline that separates a durable report from a well-reasoned guess. The reader must be able to reconstruct your conclusion without calling you — and without taking your word for anything. Below I walk through what a verifiable basis of comparison must actually contain, and how to structure it so the report holds all the way through.

What verifiability actually requires

Verifiability means that a qualified third party can take your report, follow every step, and either arrive at the same result or pinpoint exactly where they disagree with you. That is a higher standard than “reasoned”. A justification explains your line of thinking; a verifiable basis makes it reproducible.

Three questions must be answerable from the report alone:

• Which sales underpin it? Identifiable — address or cadastral reference, transaction date and the observed price, not merely an average.
• Why these in particular? Which criteria bounded the search — geography, time, property type, segment — and which sales were excluded, and why.
• How were they made comparable? Which adjustments were made from each individual comparable sale to the property being valued, and on what grounds.

If a reader cannot answer all three without guessing, the basis is not verifiable — no matter how experienced the valuer is. Trust in the author is no substitute for a trail.

Document the source, not just the figure

A figure without a source is an assertion. Every data point in the basis of comparison must be traceable back to a named, checkable source — preferably a public one, because anyone who wants to verify it can access it.

For the factual description of both the property being valued and the comparables, you lean on the public registers, and it should be clear which ones:

• The cadastre (matriklen) for the parcel’s identity, area and boundaries.
• BBR (the Buildings & Dwellings Register) for building areas, year of construction, use and renovations — bearing in mind that registered areas do not always match the actual fitted-out area. The pitfalls of relying blindly on the register are worth knowing; we have collected them in our review of the typical errors in BBR data and how to catch them.
• The Land Registry (tingbogen) for title, charges and easements that can affect use and therefore value.
• Plandata.dk for the planning framework and use provisions.

For the price information, you refer to transaction data from the relevant market. State what you observed — the actual transaction price and its date — and keep it separate from your own derived figures such as the corrected price per m². When the source sits next to every data point, the auditor can spot-check and the bank can re-find the sale. If it does not, they have to take your word for it — and that is precisely what a bottom-of-funnel reader will not do.

Rule of thumb: If a figure in the report cannot be traced back to a named source, it does not belong in the basis — it belongs in your judgement, clearly marked as exactly that.

Make the selection visible — including the exclusions

The most common weakness in a basis of comparison is not the sales that are included. It is the sales that are missing, without explanation. When a counterparty can point to three obvious, more recent neighbouring properties that do not appear in the basis, the suspicion arises that the selection was done backwards — from the desired conclusion towards the sales that fit it.

So describe the selection as a process, not a result:

1. The search criteria — the radius, time window, property type and segment you bounded by, and why they are reasonable for this particular property.
2. The gross result — how many sales the criteria produced.
3. The screening — which sales you removed (intra-family transfers, forced sales, extraordinary terms, obvious outliers) and on what basis.
4. The final set — the sales that carry the conclusion.

When the exclusions are stated explicitly, you take the sting out of the strongest counter-argument in advance: that you cherry-picked. The craft of doing this rigorously has its own logic, which we go into in our review of how to select comparable sales for a credible valuation.

How many do you need?

The number of comparables is itself a documentation question. A price per m² derived from two sales is fragile; a single atypical sale can shift the conclusion noticeably. A broader basis dampens noise, but only if the sales are genuinely comparable — fifteen poorly matched sales are weaker than five sharp ones. We treat the trade-off between statistical robustness and real comparability separately in our analysis of how many comparable sales it takes to support a price per m². Whatever the number, the report should state it — and briefly comment on what the spread in the basis means for the certainty of the conclusion.

Show the adjustments, not just the result

No two properties are alike, so comparable sales must be adjusted before they can be measured against the property being valued — for condition, location within the area, differences in area, age and timing. This is professionally unavoidable. The problem is not that adjustments are made, but that the adjustment often disappears into a single aggregate figure, so the reader sees the end result without being able to see the path to it.

Make the adjustments legible. For each comparable sale: the observed price, the corrections you added or subtracted, their direction and size, and a short justification. Two principles keep it verifiable:

• Consistency. The same factor is adjusted in the same direction with comparable weight across all references. If you adjust one sale up for more recent condition, a correspondingly older reference must be adjusted down by the same logic.
• Separation of fact and judgement. The observed transaction price is fact from the source. The adjustment is your professional judgement. Keep them visually and verbally distinct, so the reader can see where your craft begins.

When the adjustments are transparent, a critical reader can disagree with a single correction without discarding the whole report. That is a sign of strength: a durable valuation can withstand someone poking at a single link without the structure collapsing.

When there are no direct references

Some properties have no clean basis of comparison at all — new construction in a new segment, a conversion with no local parallels, an atypical commercial property. Here the documentation becomes more important, not less. When you cannot point to direct comps, you must instead document the method you replace them with: which indirect anchor points you build on, which assumptions you make, and how sensitive the conclusion is to them. That exercise — pricing on the existing market without direct comparable sales — deserves its own discipline, which we have covered in our review of pricing a new build without direct comps. The point for documentation is the same: the further you move from direct comparability, the more clearly the assumptions must be stated.

Structure the report so the trail is easy to follow

A complete basis that is hard to navigate does not get verified — it gets skimmed. A readable structure is part of documentation quality. An order that follows the reader’s questions works in practice:

1. The property being valued and its factual basis, with sources cited.
2. The selection criteria and the screening.
3. The comparable sales individually — identification, observed price, source.
4. The adjustments per sale, with justification.
5. The conclusion, with a short note on certainty and caveats.

And an honest delimitation at the end: this is professional valuation on the existing market, not a guarantee and not legal or accounting advice. Where rules, registrations or market conditions may have changed, the relevant data should be verified in the current source. That does not weaken the report — it shows that you know the limits of what the basis can carry.

From manual trail to structured basis

Most of the work above is not technically hard — it is time-consuming and easy to cut corners on under time pressure: pulling the right sales, screening out the unsuitable ones, keeping source and adjustment separate, and laying it all out so it can be verified. This is precisely where a structured basis pays off, because the report has to stand before a credit committee or a counterparty without you in the room.

Arcili’s Ejendomshandler module (Comparable sales) is built to make the collection and bounding itself verifiable: comparable sales from the market within a radius of up to 4 km and 24 months, with the underlying sales visible and a PDF export that documents the selection rather than merely presenting a figure. It does not replace the valuer’s judgement — the adjustment and the conclusion are still your craft — but it gives a traceable starting point, so the documentation is in place from the outset.

If you want to see how this looks on a concrete case, take a look at Arcili or book a walkthrough, and we will show you the basis from search to finished export.