Welcome to the Q&A Forum

Do people tend to search for "CF-19" in the Toshiba example, or do they tend to search for "CF-1956Y6XLM"? If it's CF-19 then I would add more value to the example pages, and not worry about the subpages as much. But, I'm guessing that it's the specific model numbers, in which case the ideal situation is to be able to index an exact page for that model number. If you take a look at the "CF-1956Y6XLM" example, PC World is ranking #1 pretty much on all spec content, meaning they're coasting on domain authority to rank those pages. Meanwhile I see you guys at #4. Typically I would suggest that it's a bad plan to go with really thin content, but if everyone else is doing it, you may not need 200-300 words to move up in the rankings. Try producing 50-75 custom words on 100 of these pages where you're ranking Top 5. Do it for models that are newer so you can monitor ranking improvement over time. If the ranking and traffic improvements happen, and they convert, then figure out if you can scale that process up for every new incoming product. Other SERP benefits can beat rankings here, too. If you can get legitimate product ratings and generate some rich snippets for the products, that will help maximize your CTR. Try to write better meta descriptions, too - right now they're all pretty drab on that SERP example. Martijn's suggestion of reviews is a good start but will probably only help on 10-20% of pages that you're able to get reviews on. Nevertheless, probably worth the effort. Some e-commerce platforms will allow you to save a single product with variations, which helps with this problem. If 10 models can share a page, and be selected with a product sub menu (like the t-shirt size or color selector on a fashion ecommerce site) then that is a good way to cut down on total URLs by 50-90%. But, I'd try the unique content route first and see if the numbers add up.