Every capability sheet in the massage robot category runs the same way. Arms, programmes and pressure ranges sit in bold on page one, in a typeface built for comparison. The number that actually decides whether a stranger should lie down on the thing - where its force is capped, which standard it certifies against, what happens the second after someone hits stop - usually sits several pages back, if it is printed at all.
The bold page is an arms race
Nobody prints a modest number first. Capability is cheap to produce and pleasant to compare, which is exactly why it dominates a young category’s marketing before the category has settled on what actually separates a well-built machine from a poorly built one. The confusion this creates for a buyer trying to compare products at face value is covered in full at massage robots. What matters here is narrower: a bold capability page requires nothing to be independently verified before it goes to print, which is precisely why it tells a reader almost nothing about trust.
What a spec sheet is required to say is thinner than it looks
Read from the actual documents rather than from marketing copy, the legal floor under this category is lower than its confidence suggests. EN ISO 13482:2014, the standard covering physical contact between a robot and a person in a personal-care setting, requires a manufacturer to limit the force the machine can exert. It does not supply that number, and it says so plainly: no internationally recognised pain or injury threshold existed for this kind of contact when the standard was written, which leaves each manufacturer to set and justify its own figure rather than certify against a shared one.
The regulation that will eventually tighten this is already dated. The EU’s Machinery Regulation, 2023/1230, enters mandatory application on 20 January 2027, replacing the current directive with updated requirements for machines that interact with people. It will not retroactively fix a weak spec sheet printed before that date, but it gives a buyer a fixed point after which “we comply with the current directive” stops being a sufficient answer on its own.
The category’s confidence has outrun its balance sheet before
A capability sheet says nothing about whether the company behind it will still be operating next year, and this category has already shown why that gap matters. One operator raised $83 million in a strategic funding round in March 2025 and entered insolvency proceedings roughly a year later, in April 2026, with a reported shortfall of $157 million. Nothing on that company’s own published materials in early 2025 would have told a venue or a member to expect that outcome within twelve months.
That is not a reason to distrust every spec sheet in the category. It is a reason to treat a glossy capability page and a healthy company as two separate questions, answered by two different kinds of document, neither of which substitutes for the other. A buyer who reads only the capability sheet has answered “can this machine do what it claims” and left “will anyone still be servicing it in a year” entirely unasked, and the second question turns out to matter just as much to a venue signing a multi-year agreement as the first one does.
Insolvency itself is not the scandal here. Young categories fail companies at a higher rate than mature ones, and that is true of automation generally, not a specific defect of this one. The scandal, if there is one, is a reader assuming that funding, press coverage and a confident capability sheet were ever meant to answer a solvency question in the first place. They were built to answer a different one.
The boring answers are the ones worth trusting
The most detailed public research relevant to a force ceiling in this category was not built for it at all. Germany’s DGUV, the statutory accident insurance body, holds a register of roughly 9,000 pressure-pain measurements across 29 body locations, gathered to understand accidental contact between industrial robots and workers, not a requested session where firm pressure is the point. It is genuine, careful research, and manufacturers now lean on it for a different question than the one it was designed to answer - which is a defensible starting point only if a manufacturer says plainly which parts of that borrowed data it actually validated for a massage robot’s own force ceiling, rather than presenting the two situations as identical.
That is the pattern worth watching for on any technical file: not whether it sounds confident, but whether it answers a boring, specific question - where exactly is the limit fixed, what standard does the certificate cite, what happens in the second after the stop is pressed - in language a lay reader can follow, from a document that already existed before the question was asked. A manufacturer that answers immediately, precisely, did the work. One that answers with reassurance instead of a number did not.
Sestala’s own published numbers are printed the second way: a 70 N force ceiling, a 40°C thermal ceiling at the contact surface, a stop response of roughly 0.45 seconds, and the neck excluded from the motion path entirely, stated plainly at sestala.com/industry rather than folded into an appendix. That is not a claim that the machine is safer than any other in the category. It is what a complete answer is supposed to look like before anyone decides whether to lie down on the results - the same standard the force ceiling itself deserves in full, covered at the most important part of the machine.
