Stop backing gadget products, you twits
Book projects generally work out, and physical game projects (though there have been a couple spectacular failures) because producing books and boardgames is a mostly solved problem.
On the other hand, producing an unproven gadget is incredibly risky (as are video game projects). First, making prototypes is fun and easy. Making a production ready design and production line is anything but - suddenly you have to worry about whether that hinge can open 50K times without breaking. During R&D you just swapped out motors when one burned out - you can't do that now, all motors have to keep working. Oops, that molded plastic you used scratches really easy. Finding, characterizing, and fixing each of these things takes time. There are just dozens or hundreds of slow, tedious issues for anything you want to productize. As an engineer I've done it lots of times and it's always miserable and more work than you expected.
Second, if you get enough orders to require third world production, dealing with China is a major nightmare - I haven't had to deal with others like Vietnam but it's probably not hugely different. You might naively expect you can give them the BOM (Bill of Materials), Solidworks files, and instructions. Oh no. First you have to find someone to deal with. They will all promise you to the world. Then there will be lots of flying and calling as you laboriously explain various things and realize they haven't really understood or looked at your schematics, and they will explain to you they can't get these parts, or this part can't be manufactured like that. Lots of expensive prototypes and production test runs.
Then, and this is the worst part, even when things are working perfectly they will decide to randomly swap things without telling you. We had an entire run of printers die in weeks because they quietly changed one of the motors with a cheaper one to save and pocket 5 cents per unit. *To pocket a 1 cent saving per unit, a Chinese factory owner will happily make changes that can kill people* - I've seen it happen! They will be completely lax on quality control because they can use cheaper/fewer workers and send you batches with half the units defective. They will run your production line to make no-brand Chinese knockoffs they will sell cheaper than you. They will sell your design to other Chinese companies and prioritize them so the knockoffs are out before your product. The only way to deal with all this is to keep a manufacturing expert out there all the time to babysit - and best if they speak Chinese. Do you know someone like that?
So with all that, never back a gadget product unless it's a VERY minor refinement of an existing shipping one. And even then, why not just wait for it to be sold as a real product? So you didn't save 20%, one failed project will wipe out five of those.