MicrostockGroup

Shutterstock has "Most Popular"

iStockphoto has "Best Match"

Dreamstime has "Relevancy"

Fotolia has "Pertinence"

StockXpert has "Popularity"

They all have different names, but they are all essentially algorithms that are trying to return the most relevant results when a buyer searches for images.

Which stock site do you believe returns the most relevant results when you search for images?

And why?

I just did a search using 3 of my common keywords to see how many of my images appear in the first 20 images.

IS  - 0 
SS - 1
StockXpert - 3 
Fot - 12 
DT - 16

Now I know why IS is completely tanking for me and why fot and Dt are doing so well.
At IS my first image takes 265th place.  Before the latest best match change I always had 2 or 3 in the top 50.   The one that is at 265th place has only sold 9 times when I have other images that have sold over 1000 times that I can't even find they are now so far back in the search.

SS is actually doing well at the moment but I haven't uploaded this sort of images for a while so they are not showing as well as othes in the "best match" search.

I haven't voted because to be honest I don't think any site has or can have a search engine that brings the 'best match' results that any buyer could want.

As a buyer when I search for an image I want the best image that matches what I'm searching for, I don't care whether it's been downloaded once, fifty, or a thousand times, I don't care when it was uploaded or how succesful the contributor is on that particular site, because to a buyer all that information is irrelevent and as a buyer I and only I can make the decision as to whether the image is the 'best match' for the purpose I want.
I'm also not bothered whether an image is exclusive to a particular site because for RF exclusivity is irrelevant.

With that in mind there is no search engine or site that can possibly determine what is 'best match'.

And as a buyer I am aware of this and I'm also aware that if I really want the best match I'm going to have to do it the old fashioned way and go through the pages of results.

Personally I'd like to see more time spent working on making sure that the images that come back as a search result are actually relevant to the search terms, iStock with all the will in the world have tried to do that but to be honest the result is a disaster because their CV is useless and doesn't recognise nearly enough search terms, so if I want an image now I go to another site and search the old fashioned way.

And 98% of my time is as an contributor, for a full time buyer it must be hell.

That is exactly what the poll is trying to reflect.  Which microstock site has the search engine that gives the most accurate results?  I  believe that is what the "Best Match" is supposed to reflect.

Results don't necessarily reflect a good or bad search engine.  If people spam images and they appear in a search due to the spammed terms, the search engine is working correctly - contributors aren't.

Making no tests I would vote for DT, because their engine takes into account the title and description too.  FT's choice of most relevant keywords on top of the list is also a very good one - unfortunately things don't seem to work well there.

Regards,
Adelaide

I think that Fotolia has the most fair system and thus best in terms of exposure for all contributors. it's relevancy search is now an amalgam of new images that are yet to be sold and old solid sellers.

I think DT has the best search results, for the reasons Adelaide mentioned (title and description included) and also because they take rejection rates into account. 

Folks who consistently upload quality work are rewarded by higher search placement and the customer benefits from seeing work that is high quality.

Unlike ratings or views, which are easily gamed, the only way to game DT's search is by uploading good images 

ETA that using rejection rates is also fair because a skilled newbie has as good a chance as an established contributor.

DT!
Reasons - everything that Lisa, Hoi Hu and others have already said.

Never under estimate the skill and ability of the buyer in using the search engine of any particular micro site to find a file for a project.
"socks" "nobody" " white background" is a simple string of keywords and can be easily  typed into istock's search engine and hey presto my 2 useful images appear.
"businessman" "office" would not be good enough terms for any search engine on a particular micro site to narrow down a search for a  photo but i suspect a buyer would of learnt that along time ago.

Not a microstock, but i find Alamy's algorithms (AlamyRank) working very nicely, much better than any microstock. Keyword spammers and people who upload too many similiars do sink to the bottom of their search.