Appraising flats vs houses: where the same method needs different care

The £/sqft method works for flats — but the naive version of it fails in specific, expensive ways. Four adjustments matter.

1. Tenure and service charge drag

Flats are overwhelmingly leasehold, and buyers price the whole package: ground rent history, service charge level, management quality. Two identical flats with £1,200 vs £3,600 service charges are different products. When comping a proposed block, prefer evidence from blocks with comparable charge structures.

2. The block effect

Flat values cluster by block, not just by street — one well-run building can carry a 15% premium over its neighbour. Comp sets drawn from many blocks average away exactly the variation you need to understand. Look at the per-block spread, not just the area mean.

3. Floors, aspect and outdoor space

Within a block, £/sqft varies by floor level, aspect and balconies more than houses vary by garden. New-build schemes can capture this with unit-level premiums — but only if the base rate came from genuinely comparable stock.

4. Absorption and the investor share

Flat schemes lean more on investor purchasers, so absorption swings with mortgage rates and yield maths, not just local owner-occupier demand. Check what share of nearby flat sales were quick multiple completions (investor blocks) versus steady one-by-one owner sales — the pattern predicts your own sales programme.

Method unchanged, care redoubled: local, indexed, type-specific evidence — read with a flat-shaped eye.

Data referenced in Threshold is displayed under the Open Government Licence (HM Land Registry Price Paid Data, EPC Register, ONS, UK HPI). All figures produced by Threshold are statistical estimates for site appraisal, not RICS valuations.