I am not sure reviews can ever be profitable business. I think you have to invest a lot of time in the review itself and in getting to know the field.
A while ago, I was looking for a wireless phone charger. A friend recommended Wirecutter for reviews. I took a look and bought their favorite charger. It's great!
I then later wanted to buy a 3D printer, so used the same strategy. I bought what The Wirecutter recommended, and as soon as I got the printer, I realized that they are completely clueless. The printer I bought doesn't even let you use gcode from an external slicer. You have to use theirs, and it is awful. It adds supports and rafts to everything. It has to be connected to a Windows computer for the entire duration of the print. And it requires you to create an account to even open up their software. It is so unbelievably bad that I was blown away that anyone could like the thing. And that's the problem with reviews; these people don't know anything about 3D printing, so they opened some software, inserted a cube, printed it out, and hey! a cube! wow! But you can't actually do anything useful with that.
I ended up extensively researching everything on my own and decided on a Prusa i3, and... yup, exactly what I wanted. I enjoyed spending a day putting it together. I enjoy hacking up the firmware and flashing my own build. I enjoy the huge community of printable add-ons. That is what I really wanted. And, of course, the prints are great. (Why did I read the Wirecutter and not something dedicated to 3D printing? The reality is, everything 3D-printing-related seems like machine-written blogspam. Writing a blog post that contains the word "filament, slicer, spaghetti, and geared hotend" 100 times and getting some ads from AliExpress equals instant wealth beyond your wildest dreams. Or at least that's what Google search rewards.)
Anyway, where I'm going with this is that you kind of have to be an expert in the field of products you're going to review. You have to develop extensive test protocols, beyond what manufacturers do internally. Remember all the bad USB Type C cables? Benson had to develop his own hardware and software to properly evaluate them. And that's for a strand of wires. Intelligently testing a 3D printer for its print quality, software quality, and long-term reliability would probably take years. By the time your review came out, the product wouldn't be on the market anymore. So nobody does it, and it will never be profitable to do it. (With the Prusa printers, what gives me hope for reliability is that they manufacturer the printers using their printers. If there is a fatal longevity problem or irritating software bug, they have to fix it, or they can't manufacture any more. That is the real review.)
I used to subscribe to Consumer Reports, who does seem to do a lot of testing. But they are too far behind and their test coverage was too limited to be useful to me. I wanted to buy a vacuum cleaner, none of the brands they tested were on the radar (it was just stuff you could buy at wal-mart, nothing high quality). Same with flat-screen TVs. And to some extent, you can never know what the reader of your review really wants. If I were buying a TV, I want it to accept HDMI and Displayport and copy the pixels out of the digital stream directly to the display panel. A pixel value of (0, 0, 0) should emit no light. A pixel value of (255, 255, 255) should blind me. When I press the power button, the TV should begin displaying the input signal the instant the button makes a clicking noise. Not 100ms later, not 500ms later, and certainly not 10 seconds later like my Samsung.
Such a product does not exist, so there is no product that would ever make me happy or get a 5 star review. After that, it's just a list of compromises you can accept. (Does anyone even make a non-smart TV? I watch TV so infrequently that I have a new firmware update available every time I turn the thing on. It annoys me greatly.)