In Q3 2024, we deployed 38 electric standing desks across two floors. Within six weeks, three workstations were filing the same ticket: "The desk wobbles above halfway."
My first reaction was the frame. I pulled two vendor quotes for faster dual-motor frames — $140 more per desk. I almost signed it.
I didn't, and only because I forced myself through one boring exercise: comparing the wobbling desks to the steady ones, spec by spec.
The complaint you hear is not the real complaint
The wobbling desks were identical to the steady ones in every way that shows up on a spec sheet. Same frame. Same legs. Same 355-lb rating. The only variable: desk surface size.
The ones that wobbled were configured at 72×24. The ones that didn't were 60×30.
Same procurement batch. Same warehouse. Same installer, in fact. The difference was the price ratio between depth and width.
From the outside, this looks like a frame quality problem — mostly because the frame is the only part anyone advertises. The reality is that on every standard-tier configuration I've inspected, the frame is the part least likely to fail. The surface above it is the part nobody tests until employees start filing tickets.
The part that actually fails, in one sentence
Stability at height is not primarily a motor problem. It's a lever problem.
A desk top is a board. The wider it is, the further the mass sits from the legs. The thinner it is, the more easily it twists under its own weight and the weight of your arms. A 72-inch top with 24 inches of depth, a 1-inch core, and dual monitor arms is a 72-inch cantilever. Every keystroke turns into a wobble.
Swap that same frame for a 60×30 top at the same thickness and the wobble disappears — not because the frame improved, but because you shortened the lever and added 6 inches of depth to absorb the torque.
That's the part I want to flag for anyone doing a fully jarvis standing desk review 2025: the depth spec is not a styling choice. It's a physics variable.
The spec almost nobody compares
Fully's Jarvis line offers different depths depending on top material and width. As of January 2025, the configurator on fully.com lists rectangular and L-shaped tops in roughly 27 to 30 inches of depth depending on which top you select and which width (confirm current specs on fully.com before ordering — configurators update regularly, and pricing moves faster than reviews do).
The 30-inch option is the one that most changes what your lever actually looks like at full standing height. It's also the one that gives you room for a monitor arm and a keyboard tray without the front edge hanging over your toes.
It's tempting to think all 60×30 tops are the same thing, so you can just compare unit price. But a 60×30 in laminate and a 60×30 in bamboo can have meaningfully different stiffness because of core composition and edge construction — and that difference is usually invisible in a side-by-side chart. If it's not on the quote sheet in inches and pounds, ask for it in writing before you approve the order.
One thing that isn't in dispute: the Jarvis tops we pulled in Q1 2025 all matched published dimensions within tolerance. No surprises there. But that's the baseline, not the ceiling.
What buying by unit price actually cost us
Here's the number that made our VP of Operations stop scrolling his phone.
Vendor A quoted a 72×30 top at $420 per unit. By the time it landed, it was $542 — $44 for freight (amortized across the order), $51 for 1.5 hours of internal assembly at $34/hour, plus $27 for cable management hardware we didn't budget for. That's $542 against a $640 all-in quote from the other vendor.
So the "cheap" option was $98 cheaper per unit. Then 3 of the 38 tops had to be swapped because of wobble under monitor-arm load. That swap cost $1,840 total, and it added 22 days to our floor reopening schedule.
After the swap, the "cheap" desk was $50 cheaper per unit — before labor. That's what happens when you compare unit price and pretend the rest of the bill doesn't exist.
I knew I should have checked core thickness before sign-off. I told myself it was "the same series we've always ordered." It was. That one time was the difference. $1,840 and a three-week delay.
Look — freight, assembly, cable management, returns, and a 22-day schedule slip are not edge cases. They're the normal bill. They just don't make it onto the first page of the quote.
What actually worked (short version)
Once the problem was clear, the fix was small. Which is usually how it goes when you spend real time understanding the failure before jumping to a replacement.
We rewrote the spec to require depth, core thickness, and top weight as separate line items — not a single "premium top" marketing phrase. We pushed our evaluation to favor 30-inch depth on every top wider than 60 inches. And we now calculate a per-unit fully loaded cost before the first quote review, not after the first return.
That framework held up in our 2025 fully jarvis evaluation. The depth options are real, the price difference between 27 and 30 inches is small relative to the frame upgrades people chase, and the fully jarvis standing desk specs we requested came back with numbers we could actually verify.
To be fair, the frame matters at the top of the height range — motor quality shows up in how smoothly a desk raises and how it holds position over 5 years. I'm not going to pretend it doesn't. But between two mid-tier frames that look identical on a landing page, the thing that decides whether your desk is usable is what's sitting on top of it.
Will this make a standing desk wobble-free? No. Nothing does, from Jarvis or anyone else. Physics doesn't negotiate with marketing.
But a sensible depth ratio, a top heavy enough to fight torque, and a total cost that accounts for the parts of the bill nobody prints — that's what turns a standing desk from something you regret into something you forget. Which is the whole point.