In the short term, I don’t think that Purism can pursue any other strategy except go for the low-volume, high-priced niche market, where people are willing to pay a premium.
Purism could sell it its laptops for cheaper if it switched from custom manufacturing to Clevo base models, like most Linux laptop companies do, but then you don’t get hardware kill switches, CPUs fused to allow an unsigned BIOS/UEFI and TPM chips. For a Clevo model laptop, people wouldn’t not be willing to pay a premium, so how do you pay for 3 developers of PureOS and 1 developer of Coreboot, when you are selling at standard prices?
System76 managed to get to high enough volume that it can afford to develop its own distro and pay for a Coreboot developer, but it took System76 12 years to get big enough so that it could maintain its own distro and 13 years till it could hire its own Coreboot developer.
Just to make up some numbers, let’s guesstimate that System76 sells 5 times as many laptops as Purism, so System76 can charge a $100 markup per laptop, whereas Purism needs to charge a $500 mark-up just to employ the same number of software developers to maintain their distros and Coreboot ports.
With the Librem 5, I have often wondered it Purism shouldn’t have been less ambitious in version 1 of the phone that it could have sold for less. For example, Purism could have done this:
Version 1 for $400: Snapdragon SoC and UBports’ Ubuntu Touch (no hardware kill switches, no lifetime updates, no RYF certification, and no open schematics because of Qualcomm’s copyright over reference designs),
Version 2 for $500: NXP i.MX 8M Quad with hardware kill switches and UBports,
Version 3 for $500: NXP i.MX 8M Plus with hardware kill switches, PureOS/Phosh and OpenPGP card slot
In retrospect, it would have been much less risky and probably a better business plan for Purism to have implemented the Librem 5 in three versions, and it would have avoided the public relations nightmare of being 19+ months behind schedule. However, Purism couldn’t have charged much for version 1 of the phone, because it has no compelling features that make it better than a Sony Xperia with SailfishOS. It also would have meant pouring a lot of investment into UBports, which is is a huge code base of siloed tech and is going to be hard to maintain over the long term, so you can’t promise lifetime software updates. Then, you have to abandon all that investment in UBports for something more maintainable and you are going to piss off a lot of users who have grown accustomed to UBports.
I honestly don’t know if Purism used the right strategy in deciding to implement everything in version 1 of the phone at a very high selling price, rather than implement the phone in stages at a lower price that would have generated more volume. The issue is that Purism would be operating with planned obsolescence and would receive a lot of criticism. The people who invested in versions 1 and 2 would be angry that their platform had been abandoned after a couple years.