I recall that years and years ago many contributors to iStock were complaining about 'good days followed by bad days'.  I experienced it myself - a couple of really good days would be followed by a sudden bad day.  It all smacked of income limiting and was very obvious at the time.

But they didn't do it by manipulating the search.  Search positions remained stable but sales dropped significantly.  It seems the way they did it was to limit what registered buyers could see - in other words the search results seen by individual buyers were not necessarily the same as the results seen by contributors or visitors.  iStock themselves confirmed that they could 'tailor' the search results for all registered buying accounts.

And this is probably what is happening at Shutterstock.  Yes the search might move around a bit generally, but if they really want to control sales they simply have to remove or reduce certain results seen by particular buying accounts.  They know which accounts are subscriptions, on-demand etc, and they know which contributors have had a run of above average sales or SODs.  It would be a simple process to pull back those contributors from visibility to certain buying accounts.  None of this will ever be seen by contributors or visitors - only by the registered buyers.

Interestingly I've noticed a similar situation on Dreamstime.  With their famous 'three good weeks followed by two bad weeks' again the search doesn't seem to change, yet sales can be dramatically different.  Again I suspect they are able to tailor results to particular buyers accounts.  For instance in January I had a very good month of on-demand sales.  Very good.  For the last couple of weeks though, all I've had is subs.  Somehow the on-demand sales have completely disappeared, more than just coincidence.  They only way they can do this is to 'remove' my portfolio from the registered on-demand customers temporarily.  Of course only the customers will actually see the changed search results.

Your post should get more upvotes. It's the simplest "non-conspiracy" explanation and like yourself have noticed, it's not SS exclusive. I have tried searching for my images using proxies but they yield the same results. All non logged users see basically the same all over the world. It would be interesting to see the experience of contributors who also have buyer accounts.

So my idea, which I;d do myself, but I'm pretty rubbish at Excel... is to create a kind of calculator. Here's your boxes:

- Sales for 2016
- Average number of items you had in 2016
- Average number of items you had in 2017
- Average number of items SS had in 2016 (video number if you just do videos, photo if you just do photos, combined if both)
- Average number of items SS had in 2017 (video number if you just do videos, photo if you just do photos, combined if both)
- Expected sale for 2017 (based on 2016 sales multiplied by increase in your content, divided by amount of new competition)
- Actual sales for 2017
- Difference between the two

And that last one is the big one. We'll call that X. You see, currently, everyone has different amounts of items, that sell better or worse than everyone else... and everyone is uploading new items at a different rate, so it's really hard to make comparisons. So people's experiences, rightly or wrongly, come off more as opinions rather than facts. Mainly because nobody really wants to, or can be bothered to, share all their data.

With this calculator, all you have to do is share that last figure, hopefully over multiple years. So (for ease of maths) if you make $1000 from 1000 items in 2016 (And SS have a total of 10,000 items), and you have 1500 items in 2017 (And SS have a total of 20,000 items) then your expected earnings would be $750. So if you make $375 then X is -50% and if you make $825 then X is +10%.

I could be wrong, but I've got a vague feeling that X might be pretty much the same for most authors. Give or take. Any increases over you expected earnings could be explained by more exposure to your items over time, for example... and any decreases could be attributed to people already having bought your item or it not being as relevant any more. I/e/ if somebody sells mainly images of people holding mobile phones, they're going to notice a bigger drop in older items than somebody who does photos of landscapes and the like.

If everyone's percentage increase/decrease is around the same, then it's very likely that there isn't a cap. If the results are all over the place then it's possible there is one. I just think there's too many variables for any comparisons to be drawn with people just giving anecdotal evidence. A lot of times you'll get stuff like this that people take at face value when there's a bit more to the story...

"I can't believe I've only had 5 sales this month!". Yeah, but you normally get 6 a month, so it's not all that crazy.
"My sales have dropped 50% in the last year!" Yeah, but did you forget you deleted half of your portfolio this year?
"Sales have steadily been decreasing year on your". That's understandable since you haven't uploaded anything in five years.

So yeah... if anyone wants to make the SSFSSSC* then that would be great. Then there's no "ooh, but this", or "ahh, but that". We can say... "what's your X%? Give us the numbers"

*Space Stock Footage Shutterstock Sales Calculator   

There's lots of things it doesn't account for. Unfortunately it's not a grand unified theory of everything to do with Microsoft sales... just hat it means there will be a point of reference when comparing increases and declines.

At present there's far too many variables to draw any kind of useful data from peoples' anecdotal evidence of how they are doing. This would still leave variables, but it would eliminate the big ones... like how much somebody normally makes, how much new stuff they have uploaded over the past year, and how many more files they're competing against. So theres a bit more of an even playing field when discussing performance.

Also the "cap" theory is that it applies only to the top 1% or whatever the figure is therefore the evidence that 99% don't get a cap doesn't prove anything.

There seem to be several types of cap under discussion.

One is "trust me I've been doing this a long time" cap for earnings. Apparently the little people will never suffer this.

Another is a short-term thing: once you've earned so much in a week or a month you earn little or nothing more during the rest of  period

Another is the good-day-followed-by-several-bad-days cap.

I just go by the heuristic principle otherwise known as Occam's razor. If there are easier ways to explain something, in the absence of evidence, there is no need to complicate the explanation. On the other hand, I understand the frustration, and if a belief, however random, makes it more bearable, go for it.