Journal · Industry data
What Which?’s 2026 kitchen tests tell you — and what they can’t
Which? bought 21 kitchen ranges from nine big brands, put them through 19 lab tests and asked 4,580 owners how theirs are holding up. The spread — 89% to 54% in the lab — is the most useful number a kitchen buyer will see this year.
Most kitchen marketing is adjectives. Which?’s 2026 kitchen-brand assessment is the rare exception: the consumer group bought 21 ranges from nine of the biggest names — Howdens, Magnet, Wren, B&Q, Ikea and Wickes among them — at market price, sent them to a specialist laboratory and ran 19 tests across the quality of the units, the loading of shelves, the drawers and runners, and the hinges. The best range scored 89%. The worst scored 54% — a 35-point spread between products that look, on a showroom floor, much the same.
Alongside the lab work, Which? surveyed 4,580 owners in February and March 2025 about satisfaction, value and durability. The brand-by-brand scores sit behind Which?’s paywall and belong to them; the aggregate picture is public, and it is the aggregate picture that matters if you are about to commission a kitchen from anyone — us included.
The headline numbers
ranges from nine brands bought at market price and lab-tested
between the best and worst lab score — 89% against 54%
of kitchens showing damage or wear after five-plus years of ownership
the customer-score range across the brands surveyed
It fails at the hinges first
The two lowest-scoring ranges failed the hinge-strength test outright — weight applied to the open door, the one moving part every cabinet depends on. The owners’ survey tells the same story from the other end: at five to ten years, the most common problems were scratches to the cabinet finish (15%) and soft-close mechanisms failing (12%), and 32% of kitchens had needed a repair of some kind. Even the most reliable brand had 34% of owners reporting wear, against the 44% average — better, but not immortal.
If that pattern sounds familiar, it should. When we read Which?’s fitted-bedroom survey earlier this month, the same two culprits led the damage table: the finish and the drawers. The parts of fitted furniture that fail are the parts that move and the surfaces you touch. That is also precisely where money gets saved in a cheap carcass, because it is where the saving is invisible on day one.
The parts that fail are the parts that move and the surfaces you touch.
What the lab cannot load-test
The method is honest about its edges, and the edges are worth knowing before you lean on the results. Prices were mystery-shopped in May 2025 against a standardised 12-unit kitchen — a sensible way to compare chains, and nothing like the way a bespoke project is priced, where the survey and the layout drive the number. Independent kitchen companies appear only as a single averaged category, because no one independent produced a big enough sample to be scored — so the tables can tell you a great deal about the chains and, by design, almost nothing about any particular workshop, good or bad.
And a laboratory can hang weight on a door and load a shelf. It cannot test the things that decide how a kitchen feels in year six: whether the survey was accurate, how deep the installation went, and who answers the phone when a soft-close stops closing softly. The customer-score spread — 80% down to 65%, with two-star service ratings at the bottom — suggests the difference between suppliers is as much about what happens after the template as before it.
What we’d take from it
Three questions fall straight out of the data, and they are fair questions to put to any supplier — including us. First: what are the hinges and runners, and who makes them? The two lowest lab scores failed on hinge strength alone. Second: what is the finish, and how does it wear? A scratched finish is the single most common complaint at five-plus years, and a finish fails at the edges you touch first. Third: what happens after installation? Ask who comes back, on what terms, and for how long — the survey says that is where brands separate.
The honest caveats: the survey population skews towards the big chains, the brand-level detail is Which?’s to publish and not ours, and none of these numbers describes any individual independent maker. What the test proves is narrower and more useful — kitchens that look alike in a showroom are not alike under load. The questions above are how you find that out before you sign, rather than in year six.