How We Pick

Start with the uncomfortable part

Products we have put on a test rig and measured: none. We do not have a lab, a decibel meter or a deflection gauge, and a lot of sites in this category imply that they do. We would rather lead with that than bury it.

Of the 24 products on this site, 1 is owner-tested and 23 are researched. Those are different claims, and every product carries its label on every page it appears on — in the quick picks table, on the product card, and in the comparison tables.

What the two labels mean

Owner-tested

Bought with our own money and used daily. Exactly one product qualifies: the UPLIFT V2 that Mike works at. Even there, we do not publish measurements we did not take — no decibel readings, no wobble figures, no month-by-month wear log invented after the fact.

Researched, not personally tested

Chosen the way an informed buyer would choose: read the maker's specification page, the manual and the warranty terms; record the figures that decide whether a product fits a body and survives a decade; check whether replacement parts are actually sold; then score it against the rubric below. Every specification carries a link to the page it came from.

The rubric

Each category has five weighted criteria. The weights are chosen to reflect what decides long-term use rather than what photographs well, and they are visible on every product card as a scored breakdown rather than a single opaque number.

Standing desks

• Height range, 30%. Measured at both ends. The floor of the range disqualifies more buyers than the ceiling does.
• Stability at height, 25%. Leg stage count, frame geometry, published capacity relative to a realistic load.
• Warranty scope, 20%. What is covered, not how long. Electronics coverage weighs more than frame coverage, because frames do not fail.
• Capacity headroom, 15%. Rated capacity against a typical loaded desk.
• Parts and support, 10%. Whether a control box can be bought in year four.

Everything else

The same structure with category-appropriate criteria: belt length against stride for walking pads, lift range and desk footprint for converters, lumbar adjustment and warranty for chairs, load capacity and whether an arm holds position for monitor arms. The exact five and their weights appear on each product card.

Scores are editorial judgments from documented research, not lab results. A 9.5 means the published specification is excellent for its purpose, not that we measured it.

Claims we refuse to make

• No invented reviews, testimonials or quotes. Not one appears on this site, and none ever will.
• No star ratings or aggregate ratings. We do not collect them, so we do not publish them — and we deliberately emit no Review or AggregateRating structured data, which would claim ratings we do not have.
• No prices in the prose. Prices change constantly. Every price on the site comes live from the retailer; where we hold no verified live figure the button says "Check price" rather than quoting a stale number.
• No estimated specifications. If a maker does not publish a belt length or a dimension, the page says "not published" and names it as a mark against the product.
• No health claims beyond the sources. We cite UT Southwestern and a Cochrane systematic review, including where Cochrane finds the evidence weak. The honest version is here.
• No medical advice. Nobody here holds a clinical credential, and the health pages say so on the page rather than in a footer.

How we are paid, and what it does not buy

Deskstance earns affiliate commission when you buy through our links, at no extra cost to you. Nobody pays for placement, no brand sees a page before it publishes, and no commission rate is consulted when a ranking is set.

The clearest test of that is what we recommend against. We tell you the value pick often beats the premium one, that a converter is sometimes a better buy than a desk, that some readers should buy nothing until they have measured their elbow height, and that a discount claimed by phone — which will not credit us — is worth asking for. The full disclosure.

Corrections

Specifications change, products are discontinued, and we make mistakes. If you find an error, tell us and we will fix it on the page where it appeared and note what changed. Corrections are logged where the error was, not on a separate page nobody reads.