Anyone that reads these boards on a regular basis (especially newbies, including me at only 4 years) must be puzzled by the wildly different results we all seem to have from each other. Just look at some of the posts. "DT sales dead", "SS dropping badly", "BME on Alamy".
OK...I made that last one up; but you get the idea. There are always posts that people make giving totally different experiences. We can all have different sales within the same site and wonder if it's due to management, policy changes, partner sites, etc.

But I'm thinking there might be another reason that we don't think of or check: what are the subjects of the photos? It seems to me (looking at my own portfolio) that sales of the same image can vary wildly from site to site. Why? I dunno. Different types of search technologies? Attracting different types of art directors? Attracting more bloggers than designers? Who knows?

So, here's an example of one of my images. Both uploaded the same day.
http://www.shutterstock.com/pic.mhtml?id=200092391

Total Sales on SS: 9
Total Sales on FT:  331

Hard to know just what to do with this information. But sure is interesting.

My point of view as a new footage contributor (less than a year) is that microstock isn't a exact science. There are many variables there between content quality, quantity, variety, keywording, competence, the marketplace it self, search engine and the customers interests and needs.

For instance, every month my bestseller marketplace varies, from SS, to Pond5, then FT, then back to SS. Then there are those middle tier marketplaces that some months perform well and another I have 0 sales.

Also I notice, reading comments at this forum, that footage differs a lot from photos. Footage sells less than photos but some times with better prices (only some times).

This pass week I had no sales at all at any platform, while two weeks ago I had more than 15 sales in different platforms.

I agree that the only almost secure thing we can expect is our own creativity, intuition about what will sell , failure, and hard work.

Your image must have had a better place in Fotolia than it had in SS with the given keywords >> Customers tend do choose among the first couple of pages when they search for a subject.

What determines a good placement in the search? 1) Keywords 2) Search engine formula 3) Luck!

Whenever I have an image somehow manages to have a foothole in a good place I prefer not to send any similar ones to keep the first one's position.

Also subject wise diversification is very important. Different subjects sell in different times of the year. Spring photos will likely to sell more during winter whereas Christmas will sell during fall. To be able to get consistent sale numbers throughout the year one should have a portfolio consist of main subjects in the main agencies.

Before, when the libraries weren't so big, your high quality, highly commercial images would gain sales almost right away and rise to the top fairly quickly.

I wish I had started this whole stock-thing 5-6 years ago... 
(I've just started in 2014 January)

I have an actuarial background so I am heavily influenced by my past. I don't think it is possible for anyone to draw any conclusions based on their ports on any agency because each or our ports is such a tiny percent of the whole collection. It makes our experience with our ports statistically insignificant. I think it makes sense to spend time looking at the characteristics of successful images on each site and shape our uploads based on those trends.