I think the main consideration would be how unusual an image is / how difficult it would be to shoot similars, therefore being able to undercut it on the same site.

Also note that prices rose sometime between yesterday and today (at least exclusives did).
http://www.istockphoto.com/forum_messages.php?threadid=343279&page=1#post6661715

@ShadySue: that's a good point. Offering a more niche product will allow you to push higher prices through.

But since it is not always possible to objectively identify whether an image has or has no competition, or whether it is niche or not, the suggestion is to base the decision on an easier to measure metric: revenue per sale. That's whether a file usually sells in large size or in small size. If it usually sells large, then you know that the file targets bigger budget customers which could cope with a price increase.

By using that method, you can focus on the numbers you have, which are measurable, and ignore all other aspects which are difficult to measure. Obviously, there is no 100% certain strategy, but I believe revenue per sale is quite a close and objective metric for this decision.

@jsnover: you are right that E+ is riskier than P+, since the price increase is very steep. But that should not put anyone off putting the files with best chances in those collections :-)

All very good points :-) A lot has been said on whether people should participate in those collections and the impact it can have on their sales, on customers and on the industry. There are lots of opinions on whether to participate or not.

However, a few people have asked us at Stock Performer what we recommend them, to best add their files to E+/P+. So assuming you want to add files to E+/P+, what is the best strategy? Our blog post aims to shed some light on that issue and that is how we arrived to our premise.

To answer your first paragraph: If a file regularly sells large, then as you say it is being used for print. One cannot assume the customer will tolerate a price hike, but one can assume that the file does attract more revenue than others. So if I would recommend a price hike, than I would do so on those files, and not on the other ones.

But % wise, the biggest difference is with smaller sizes. They also sell the most. I feel pretty good knowing my XS files sell aat double prices, even though moneywise it makes a smaller difference than at any other file size

Especially given that today's printed photo might be wrapping tomorrow's chips, but photos on websites can be there for years.

To answer your first paragraph: If a file regularly sells large, then as you say it is being used for print. One cannot assume the customer will tolerate a price hike, but one can assume that the file does attract more revenue than others. So if I would recommend a price hike, than I would do so on those files, and not on the other ones.

Fair enough, it's one way of deciding.
Alternatively, you might choose a file that's being sold many times, no matter at what size, as it's clearly a popular file, probably without any near rival/s. Someone who buys small sizes might have a good budget, but use images primarily online.