Radius and Timeframe for Comparable Sales

Two neighbouring properties are not a data set. A whole neighbourhood with a different character is not comparable. Between those two extremes lies the professional judgement that decides whether a valuation rests on something solid or simply drifts. Every time you set the radius and timeframe for comparable sales, you pull two sliders in opposite directions: widen in space or in time and you get more transactions — but you simultaneously dilute the very similarity that makes those transactions usable in the first place.

This is not a technical setting you configure once and forget. The right radius and timeframe depend on where the property sits, how often the area trades, and how homogeneous the building stock is. An apartment in central Copenhagen and a disused farm in rural West Jutland call for entirely different search windows — and applying the same setting to both is one of the most common sources of skewed price-per-m² figures.

Why radius and timeframe are two sides of the same trade-off

Both sliders address the same underlying problem: you rarely have enough transactions that are close enough to your property — both geographically and in time — to form a robust average. So you widen the search. The question is never “what is the right radius?” in absolute terms, but “how far can I widen before the transactions I pick up stop resembling my property?”

That trade-off matters because the two types of error pull in opposite directions:

• Too narrow a search yields too few transactions. An average of three or four sales is vulnerable to chance — a single atypical transaction can move the price-per-m² noticeably. That is the focus of how many comparable sales actually support a price per m².
• Too wide a search yields transactions that are not comparable. When you stretch the radius across a motorway, a main road, or a boundary between two price levels, you blend markets that do not belong together.

The point is that the two sliders compensate for each other. In a heavily traded area you can keep the radius small and the timeframe short and still have ample material. In a thinly traded area you have to give ground on at least one of them — and then it becomes decisive which one you choose to relax.

Radius: proximity beats volume — but not always

Geographic proximity is the single strongest indicator of comparability, because it captures a whole range of factors at once: school district, the character of the neighbourhood, distance to coast and station, the noise picture, amenity value. Two properties 200 metres apart typically share most of this. Two properties four kilometres apart rarely do.

That is why the starting point is always the smallest possible radius that produces sufficient material. But “sufficient” depends on the context:

Densely built urban areas

In apartment buildings in the larger cities, transaction frequency is high and the building stock is relatively uniform within a neighbourhood. Here a radius of a few hundred metres — sometimes under a kilometre — can provide a solid basis. The short radius is a strength, not a constraint: the closer you stay, the more the references resemble your property.

Suburbs and smaller towns

In single-family-house neighbourhoods the frequency drops, and the building stock becomes more mixed in age and standard. Here the radius typically has to be widened to a few kilometres to gather enough uniform transactions — but beware the invisible boundaries. A radius that looks neat on a map can cross a parish boundary, a municipal boundary, or a divide between two price levels that the market treats in practice as two separate areas.

Thinly traded areas

In the countryside and in less actively traded areas, the circular radius often breaks down entirely as a method. Months can pass between relevant transactions, and a radius large enough to catch a handful can cover landscape of wholly different character. Here it is rarely the radius alone that saves the basis — it is a combination of a longer timeframe, acceptance of fewer but carefully selected references, and a more qualitative assessment of each individual transaction.

Rule of thumb: start as narrow as possible, and widen the radius only step by step until you have enough comparable transactions — not enough transactions in absolute terms.

The circular radius has an inherent weakness, by the way: it draws an abstract circle, not a market. A neighbourhood is rarely round. Polygon search — where you delineate exactly the area that genuinely shares character with your property, and let water, infrastructure and physical barriers define the edge — captures the market better than a circle that blindly cuts across a stream or a railway line.

Timeframe: freshness versus sufficiency

The timeframe follows the same logic. A recent transaction is more relevant than an old one, because it reflects the market as it is now. But shorter time windows yield fewer transactions — and in thinly traded areas a six-month window can produce zero usable sales.

That trade-off depends on how stable the price level has been:

• In calm markets, older transactions can still carry weight, because the price level has not moved significantly. A window of up to two years can be defensible.
• In markets on the move — where the price level has shifted noticeably within the past year — the timeframe should be tightened, and older transactions weighted lower or excluded. A transaction from a different price level is not comparable, no matter how close it sits geographically.

Note the difference between weighting and projecting forward. It is professionally sound to let more recent transactions weigh more heavily than older ones within your window. It is something else entirely to project prices into the future — a valuation should always rest on the existing market, not on a guess about where the market is heading. As long as you stick to actual, completed transactions, you avoid building on an expected price level that does not yet exist.

How the two sliders work together in practice

In practice you rarely adjust only one. The principle is to relax whichever slider costs the least in comparability for the specific area:

• In a dense, uniform urban area: keep both tight. Small radius, short window, plenty of transactions.
• In a suburb: widen the radius moderately, keep the timeframe short if the market is moving — or give a little ground on both if it is calm.
• In a thinly traded area: widen the timeframe first (transactions age more slowly in character than they spread out geographically), and widen the radius only if you can do so without crossing a character boundary.

Wherever you land, the same overarching principle holds: radius and timeframe only set the gross basis you work from. The selection itself happens afterwards, transaction by transaction. Two properties within the same radius and the same window can still be incomparable if one is newly renovated and the other dilapidated. That is why the search window is never an answer in itself — it is the groundwork for the actual selection of comparable sales, where you assess what genuinely makes two properties comparable on area, condition, age and location.

Bear in mind, too, that the public data on a building’s area and character — which you draw from BBR (the Buildings and Dwellings Register) — has its own pitfalls that can propagate into your comparison. An up-to-date grasp of the typical pitfalls in BBR data is part of judging whether two transactions really describe the same kind of building.

When the manual trade-off becomes a single lookup

Pulling the two sliders correctly is a matter of experience. But the data gathering itself — fetching every completed transaction within an area and a time span, filtering out the irrelevant, and seeing how many usable references there actually are — should not cost hours in spreadsheets and scattered register lookups.

That is precisely the exercise that Ejendomshandler (Comparable sales) in Arcili automates. You delineate the area — with a radius of up to 4 km or a freer polygon search that follows the market’s real boundaries rather than a circle — and a timeframe of up to 24 months, and you immediately get the completed transactions in the field with area, price and date. Then you can spend your expertise on what actually requires judgement: assessing how far the basis carries, and which transactions belong. A combined extract can be exported as PDF to document the valuation.

If you want to see how the radius-and-time trade-off looks for a specific area, you can explore Arcili or book a walkthrough, and we will show it on a property of your choosing.