Coming back to the original topic, I suspect that to be successful commercially in photography, you would need to be fairly ruthlessly focused on what the client wants. I don't see an issue with this; its no different, for example, from the people who make a living from mass producing commercial artworks for hotels, business lobbies etc. If you do a shoot directly for a client, then clearly you have the opportunity to discuss and negotiate, and to put forward your vision and style in your proposals. I don't know anything about the world of stock photography, but as I understand it you produce a range of product images for people to buy, but you don't necessarily have any direct contact with the clients, which makes it hard to measure the market. Equally, I suspect the stock image business holds its sales data very close so you will never know specifically what sells and what doesn't. Perhaps then, the best option is to produce a variety of different products in different styles, and see how you go. This could include more 'landscapey' pictures with more saturated colours, and more landscape features, together with more colour accurate close up pictures in more 'portraity' styles. Some clients may be looking for low key moods, others may be looking for impact, drama colour, contrast and movement. I expect that impact generally wins out, because the vast majority of photos probably have a very short shelf life. Lets face it, how many of us linger over the photos in brochures, training pamphlets, commercial websites etc where these types of things are used. Equally, those looking for photos from the millions in the stock archive probably give each image no more than a few seconds scrutiny before making a decision. However, unless you know someone who knows the business and can give specific advice, your best bet might be to just try different things. As I say, I'm not basing this on any direct experience, though!
The debate about accurate colour is also an interesting one. For me it is a very elusive subject because regardless of technical definitions and measurements, colour is ultimately all about perception. I don't mean perception in terms of I like this and someone else likes that, I mean perception literally. I think people process colours differently. Certainly machines process colour quite differently from each other, of course, and colour on screens is different from colour in prints as we know. In addition perception of colour can look different if you see a single image or two images together. With multiple images it is much easier to judge colour as 'better' or 'worse' (compared to x, and according to your taste). I am not sure where this takes us, but I suspect that subtle colour differences matter little to most people who will be viewing our photographs. While I can see differences in the rosellas in the earlier post, most of these seem to be more about sharpness and contrast than colours, and also if you saw either image in isolation I suspect you would not know that they were processed differently. The differences between them are less, probably, than the differences between different viewing devices (my main calibrated computer screen, my iPhone and my iPad all show subtle colour balance differences when viewing the same photo, which are obvious when you see them together but likely irrelevant for most viewers. All are capable of displaying 'attractive' detailed images). This is not to say that the quest for accurate colour is not a desirable and worthy one, but should perhaps be balanced against (or at least, mindful of) the way most people use and view our images.
Just a vew thoughts.
Simon