It might help if you told us what kind of images you specialize in....My experience has taught me that my particular subject matter fits better with some agencies than with others.  DP and SS are my best sellers. DT and especially 123rf, aren't. DT however always seems to come through with credit sales in the $2-$5 range so one DT sale is equivalent to 6 or more DP sales. And on the flip side, DP always seems to sell images that otherwise languish untouched by other agencies. It takes time to explore various agencies and discover which ones best suite your portfolio and which ones don't.

     DP and SS are my best sellers. DT and especially 123rf, aren't. DT however always seems to come through with credit sales in the $2-$5 range so one DT sale is equivalent to 6 or more DP sales. And on the flip side, DP always seems to sell images that otherwise languish untouched by other agencies.

It is amazing how different results are.  I earn close to 10x on DT and SS compared to the likes of DP, CS, etc. and 123 is at least double DP for me... 

As for the original question: a lot depends on how much you upload, how long you've been at it, how you keyword across agencies, whether you sell photos or illustrations, concept or landscapes, in demand or niche, etc.  Way too many variables to generalize how to do well on a particular agency!

I'd like to increase downloads but DT won't let me. As a consequence of their arbitrary "similars" policy and the latest one "Image subject is too specific or niche-oriented" my port there has about 45% fewer images than at SS/IS and others.

Earnings have increased YOY but significantly less so than elsewhere, putting DT in 5th place on my income scale and in danger of losing that position. But that's my story and it seems everyone has a different take. I will continue to submit as and when - but they have ceased to be a priority in my workflow.

Looking at the latest images it seems that DT has relaxed their similars policy. Have you tried submitting multiple takes from your photo sessions lately?
http://www.dreamstime.com/new-stock-photos-images

Thanks for the tip.... just took a peek and after looking at three pages of bugs either dead or having sex, plus various images of squished animals I'm prepared to take my hat off to the reviewers.

Exclusive images.... is it worthwhile or a money loser?

IMO it's definitely wasted money.

Exclusive images.... is it worthwhile or a money loser?

Tough to justify these days.  They only offer 7c more (20%) per download on subscriptions and 10% more on other downloads types (credit and the $2 subs).  There would have to be an incredible boost on search placement to make that worthwhile, and DT doesn't have the volume to have search placement help that much.  That said, if it is a file you ONLY sell on DT it is worth considering... 

It's good to remember that DT penalizes you for having lots of keywords. All things being equal, images with more keywords appear lower in search results at DT. When submitting there, I usually delete a few of my less important keywords. I think it helps.

I know that DT allows for a maximum of 75 keywords. Could you please tell me what will be the maximum keywords to have based on your experience ? Thanks !

My earnings have become much more erratic there lately. RPD and total downloads were pathetic last December, but Jan was good. The earnings graph is looking more like a mountain range than hills (not because of total height, but because it jumps up and down so much). I think it might be because of their on/off sales. If my port is on for the weekends and xmas break - my sales are horrible. If the non subs sales are on during a regular week, the sales are pretty good. If only subs are on during the week, the sales # is ok, but the RPD is horrible. I think I had a run with 18 subs out of the last 20 sales in December. Not surprisingly the RPD was horrible.

Yes, fewer keywords are better at DT. This is an old video from Leaf but still valid in which this admin admits that fewer, more relevant keywords help search placement.

http://blog.microstockgroup.com/keywording-secrets-from-dreamstime/

Thanks cuppacoffee ! As per the DT QA admin in this video (which I remember seeing before but could not remember the details), a good keyword count would be between 25 and 30.

DT doesn't work for me either. No matter how many images I upload to DT, sales just won't go up. Then there is the missing batch editing feature and silly categories. And their search by "relevancy" somehow always prefer my average images while the best sellers always end up hidden in the back. In fact, most of my best sellers have 0 sales on DT. I'm certainly not on the same wavelength with DT 

For not being an image buyer, I am not very good in searching, but I found the following problem at DT:
When doing a search with a result of more then 10.000 images it is impossible to see image 10.001 and further.
You get the message: Searches are limited to 10000 results, please refine your keywords for better results.
But looking at my sales page people mostly use two and only sometimes three keywords for a search.
Even then you often get more then 10.000 as a result:

City night scene: 34000 photos
valentine hearts red: more the 12.000
green hills summer: more then 12.000
mountains snow winter: 40.000
landscape mountains summer blue sky: more then 34.000
business woman card: 14.000
business woman phone: more then 38.000
yellow flower on white background: more then 62.000

I searched with �most relevant�. Using �latest uploads� was too terrible! 
Try: yellow rose on white background Only photos and Latest uploads. Take a look and enjoy!)

And when clicking on some by DT suggested keyword combinations:
�two men using tablet computer creative office� you get 4 results, in two of them is a man and a woman. And nothing creative to be found in the other two (similar) images. So too much keywords doesn't do the job either.  (This was only for fun)

Am I seeing it wrong when I think that a lot of images that are not seen at the moment of acceptance propably never ever can be found by buyers unless they show up by accident in a search or in an unusual combination of searchwords? This problem is only growing with the growth of the database.
Perhaps that people more experienced in searches can tell if I see this wrong and it can be done in another way?

I did the search for the yellow roses on white background - amazing the keywording skills of the newest uploaders!

What I thought was funny was that the site suggested related searches which were "sexy woman maid dress stockings", "african american elderly", "nuns", "Influenza" and "Pouring Jug"

We used to think we had issues with the similars search on Symbiostock, but it hardly compares to those suggestions.

Steve