MicrostockGroup

Interesting.  I'll have to try that, although I already know mine will look about the same.  I had a huge surge in November 2006 (doubled my sales in one month) and haven't made any progress despite more than doubling my portfolio since then. 

SS have 3 million images now but a high proportion of the older ones don't make much money.  They should come up with something to bring these images back to life.  Perhaps they should let us mark 5% of our portfolios as new each month.

So Hatman, are you feeling any different about marrying Istock with a potential Getty divorce?

But istock is a small part of Getty and is taking away business from the traditional site, so they might not get a good price for Getty if they build up istock.

To understand Shutterstock's sales, and why they decline so drastically after the "newness" factor wears off, you need only to look at their Search.

Yes, designers can choose "Random" as a search but I don't know why they would.  So you have a "Sort by New" and a "Sort by Popularity."  We all know what Random and New is, so how do they sort by Popularity?

I've sorted my own portfolio many times using their algorithm, and it seems "Sort by Popularity" is basically just DL's divided by time online.  It isn't 100% exact in my tests, but it is darn close to perfect.

So as an example in my case last week, I uploaded a picture of Strawberries.  It was immediately downloaded a few times the first day, probably by designers searching for New images.  So after Day 1, it now averages 3-4 DL's per day.  This high average catapults the Strawberries picture into the Top 100 most popular images in its category.

So now it is selling based on high placement in both the Newness sort, and the Popularity Sort. 

The image will immediately start to drift down the Newness search, but if it got a bunch of downloads early it will remain near the top of the Popularity search for a few weeks.  Eventually the law of averages catches up, and it cannot maintain such a high DL/day average, and the slow death begins.

How do you get images to continue to sell months after you upload them to Shutterstock?  Or in the Hatman's case, how do you get the hot sex to continue?

Your image is obviously doomed by 1/2 of the Search choices.  It will never be New again, and will continue to drift down in the order.  But if you sold it enough times in that first magical month, it can maintain a lofty standing in the Popularity search.  The thing is, I think you need to sell it an awful lot of times to keep it reasonably high in Popularity.

I now keep in my spreadsheet a list of images which have been online at Shutterstock for at least 3 months, and what their DL's per time online is.  This tells me where they rank in the search, and gives me a clue as to how often they will sell.  It seems to me it is important to stay above 10 DL's per month to maintain decent sales.  It takes a pretty special image to stay above 20-30 DL's per month.  Yes, I do sell the occasional oddball from way back which was never popular.  Probably from the "Random" searchers?  But those oddball sales are not dependable.

That's how I see the search algorithm, which by my data looks like the simplest one in the business.  The only way for older images to continue to sell at a nice pace is for them to be ranked very high in popularity, or for the search algorithm to eventually change.  I'd like to see "total downloads" incorporated into the Popularity algorithm as a factor.  If it is one now, it is a small one.

Fintastique-

Just out of curiosity, how many successful/accepted uploads do you tend to make at SS per month?  I had 65 in January, which was way above what I normally do.  I haven't decided how active I am likely to stay.