My monthly iStock earnings (red) were growing for the first few months of submitting until June 2008, then we can see a stagnation period until January 2009, and a distinct jump up during last couple of months.

And now green bars – total earnings from pictures submitted during each specific month. At the right (recent months) they are tiny – freshly uploaded pictures have very few sales at iStock. However, they are really big on the left side. During first 3-4 months of submitting I managed to produce a few bestsellers which dominated my sales. There is another smaller maximum in earning coming from my May and June uploads – a couple of other bestselling pictures. Then there is low period in July and August. I was submitting less pictures during my kayak racing season. My summer outdoor pictures don’t sell so well as my home studio productions.

This is a similar graph as the first one but for Shutterstock earnings. Trends in earnings are quite different from iStock.

Red bars show earnings from all sales a given month. So far, I have a nice, better than linear, trend up in Shutterstock. I used to produce new pictures at pace dictated by the upload limit at iStock, i.e., about 15 pictures per week. However, in September 2008, I started to shoot and submit much more pictures to all sites except iStock. These efforts show up. My current goal is 100 pictures per month accepted by Shutterstock and Dreamstime.

Green bars, earnings from pictures submitted at a specific month, reveal more interesting things. Old pictures (on left side) have significant contribution to my total earnings, but not so strong as in iStock.

The earnings from my first month submission look especially good. That was not even a full month. There is a rumor that these initial submissions are getting some preference in Shutterstock search engine to encourage new photographers. It may be true. However, in my case, the reason could also be different. It took me four months to get accepted by Shutterstock, so, during my first months there, I was simply submitting my best pictures shot in the previous few months.

In the middle of this graph we see green and red bars of very similar length. It means that the current sales were coming mostly from newly uploaded pictures. However, this trend is changing during last couple of months. Fresh pictures contribute less and less to the current sales. It is starting to look almost like in iStock.

I repeated the above graphs for a number of downloads instead of earnings, but they showed exactly the same trends.

Let’s look at RPI, return per image, calculated in a traditional way, i.e., earnings at given month divided by number of pictures in a portfolio at that time.

My RPI in both IS (blue) and SS(orange) reached maximum in April 2008, in about 5 months of building my microstock portfolio. Then, it started to decline. In frustration, I even stopped calculating it for my portfolios. Well, that trend was reversed during last months in both SS (October 2008) and iStock (last 2 months).

And now, a different RPI, calculated as a ratio of the total earnings from pictures uploaded at a given months to a number of pictures uploaded. You expect this RPI to grow in time as long as your pictures are selling.

The early submissions to iStock (blue bar on left) are really standing up. I still need to figure out how I managed to produce those strong bestsellers with a very limited numbers of submissions. I suppose that later I focused more on volume submissions to Shutterstock (yellow).

In the middle of 2008, RPI for pictures uploaded to IS and SS look quite similar while in the last 5-6 month RPI for SS is much higher than for IS. Of course, there is still a strong bias in SS sales on new pictures.

I derived above graphs from summary statistics copied from IS and SS websites – total earnings for each picture knowing its upload date. Therefore, in my analysis, I know only the total earnings from let’s say pictures uploaded to IS in January 2008. I don’t know how these earning are changing or decreasing in time. I would need to analyze records of each sale including the sale date, upload date and earnings. It would be interesting to run such an analysis for pictures selected by different criteria like time range of upload, keywords, or just tagged manually. Just a little suggestion for Rahul and Lookstat.