Hi PatR
It would be nice if you could put that company's name. I would like to see the products they offer. I suppose it's in military scope?. You can always learn things by looking at what others offer, thinking about implementing it in the drones I build.
I'll provide three manufacturers, Boeing, Aerovironment, and AAI/Textron, and all of them have been until the last few years military applications. At least two of them, Boeing and Aerovironment, initiated commercial operations in U.S. airspace under direct FAA oversight back in 2013. All of them are fixed wing examples but in principle the stable airframe concept is the same as it would work for classes of multirotors.
AAI became a government "Program of Record" with our government back around 1986. The platform used for that was the Pioneer, which remained in service with few changes aside from payloads and software until roughly 2007. In 2007 is was superseded with the Shadow, as system still in use today and similar in airframe design to the Pioneer. The Shadow is relatively unchanged from what it was at the time of release but has incorporated numerous payload improvements. About three years ago AAI began the introduction of the Aerosonde, a design that was bought from another company and originally developed by another company. The Aerosonde has been completely redesigned after the brand purchase was made. All their aircraft employ gas driven engines of various types.
Aerovironment has been highly successful since at least 2004 with their Raven and Puma class aircraft. The Puma has been and may still be in use for pipeline and other work in Alaska. The majority of system improvements have occurred via software, ground station, and payload upgrades. Aerovironment uses battery/electric propulsion systems.
Boeing/Insitu has been a sUAS market leader with their Scan Eagle platform, a gas engine platform that evolved from the original Aerosonde and Sea Scan platforms. Originally intended for the tuna fishing industry in the late 1990's it became military hardware after our government blocked the uncontrolled sale of UAS technology. Scan Eagle first saw military service in 2004 and has remained a reliable staple from then to the present. The airframe has received only minor alterations to improve flight efficiency, maintaining the same general size and shape since inception. It's modular design allows users to drop and swap virtually every component in time frames ranging from just a few minutes to an hour or so. An entire airframe can be assembled from modular components in only a couple hours by those trained on it. Payloads are diverse, with new payloads designed by Boeing and others introduced on a frequent basis. The Soft/Firmware is updated as needed to accept new components. As payloads much larger than the Scan Eagle can carry have been developed a second, larger platform was created to employ them, but the Scan Eagle remains highly desirable by Boeing's customers to this day. Although they have not sold millions of units the few thousand they have sold have their customers always coming back for more and their profitability exceeds DJI's.
Those are examples of how platform stability can combine with payload versatility to serve a broad market base over a long term and allow retention of existing customers while expanding that base further. A good design does not need to be discarded for another every 6 months to a year. The only reason for that is to induce people to retire something that works to buy something new only marginally better, if better at all. New platforms also introduce new problems, something that has remained a constant among multirotor manufacturers. The people that benefit the most from that process are the manufacturers. We should consider multirotor companies are not there to serve us, they are there to make money, and a constant phase out process serves them better than anyone else. If we were to look closely at system designs we would find that what changes most often are payloads and flight control upgrades. The components used to carry the payloads and power the systems remain largely unchanged. People constantly upgrading to new machines are paying as much or more for cosmetic changes as they are for actual upgrades. Change can be a very good thing but change needs to take place logically and only after full testing and validation has been completed. Once you have something that works you don't cast it aside for an unknown.