What's the difference between categories and concept words on dreamstime?
The categories that you choose for your images are broad concepts after all - business, arts etc. But the method of choosing categories encourages you to place your image in the most specific sub-category that you can. Example: Business -> objects or Business -> people. I think that the categories should be even more specific to better guide the user who is searching for that perfect image.
Concept words in your keywording, however can be as broad or specific as you make them. It is possible for example to use the broadest terms that you can - putting business, finance and metaphors all together on your images as keywords - BUT DON"T!
Concept keywording has gotten a bad rap because people tend to do just that - use the broadest terms. As I have explained in previous posts, this dilutes the search and creates problems for the entire site. You should strive to identify the MAIN concept for your image and use one main concept word. Then, drill down to the most specific terms possible that fit your image. The more specific you are, the better chance you have to be matched with the person searching for your image.
This is an example of Zipf's law (for the nerds out there who are interested). Zipf's law explains how the most common words in any language are used exponentially more frequently than the less common words. When an image bank - such as dreamstime - becomes so large that there are literally millions of images to search through, the only way to stand out and be unique is increased specificity in the language of your keywording -it helps you - and it also helps the dreamstime database.
So, continue to use categories at the most specific level. And use the tools that I have show in previous posts (The Practical Keyworder – Visual tools for Visual people)
to come up with the most specific - yet relevant terms - to describe your images. Here is a link to an interesting article on Zipf's law and how it effects search if you'd like to read further: Zipf’s Law: how can something so simple explain so much
As always - Good Luck!