MicrostockGroup Sponsors


Author Topic: Dreamstime in may - search engine changes side effects?  (Read 14619 times)

0 Members and 1 Guest are viewing this topic.

« Reply #25 on: May 23, 2009, 16:57 »
0
December was worse, but that was really really bad.  If it wasn't for a few dozen sub sales with keywords local to my my area, this would be my worst month in 2 years.  And to think 60 days ago was my BME on DT.


« Reply #26 on: May 23, 2009, 17:04 »
0
finally I get the positive of a search engine change!!! I always seem to take a hit when IS or someone makes changes. 
I'm looking to scrape in a bme for DT this month (which is a huge improvement for the past few months).  sorry for those that lose out

« Reply #27 on: May 24, 2009, 06:48 »
0
DT down for me also (-50 % compared to previous month) but IS is up so  it is fine :)

« Reply #28 on: May 24, 2009, 08:32 »
0
So this is how my stats look like at this moment. Since the search engine changes everything just died off.
Anybody else having the same kind of stats? Any ideas about how to improve?
my graph is very similar :(

« Reply #29 on: May 24, 2009, 11:01 »
0
Warren, don't bother posting anything critical on the DT forum. They're extremely thin-skinned and always watching for negative posts, which they'll just remove.  I once received a nasty warning from them hinting that retaliation is possible, so don't mess with them if you want to continue selling there. 


WarrenPrice

« Reply #30 on: May 24, 2009, 13:55 »
0
My point is that NOONE will share the information.  I wan't being critical at DT ... at least I did not think I was being critical.  I'm just being quizzical.  How do people find out about changes in the search engine?  Does DT notify the upper echelon contributors?  Is there a special forum where such things are shared?  I'm just being curious.  I have no idea how DT search engine works.  Therefore, I have no idea what to change to take advantage of searchs.  But, I sure would like to get educated.   ;D

lisafx

« Reply #31 on: May 24, 2009, 16:26 »
0
My point is that NOONE will share the information.  I wan't being critical at DT ... at least I did not think I was being critical.  I'm just being quizzical.  How do people find out about changes in the search engine?  Does DT notify the upper echelon contributors?  Is there a special forum where such things are shared?  I'm just being curious.  I have no idea how DT search engine works.  Therefore, I have no idea what to change to take advantage of searchs.  But, I sure would like to get educated.   ;D

I can't give a definitive answer.  In the absence of a big announcement, like istock's when they released best match 2, the only way I ever find out about search engine changes is if I see a big change in my download patterns.  Then I go on the site forum or this forum and look to see if others are experiencing the same.

If a large group are experiencing the same, and there are factors they have in common (like seeing more new images selling, or older images, or exclusive images, for example) then I assume that the search engine is responsible.  This can also be checked by doing sample searches yourself on a regular basis. 

Yeah, it definitely takes a lot of detective work or research to verify it.  The sites aren't talking. ;)

« Reply #32 on: May 24, 2009, 18:19 »
0
I have no idea how DT search engine works.  Therefore, I have no idea what to change to take advantage of searchs.
well of course, they're not going to reveal the details of their search algorithms. No microstock will tell you that.

« Reply #33 on: May 24, 2009, 21:50 »
0
My point is that NOONE will share the information.  I wan't being critical at DT ... at least I did not think I was being critical.  I'm just being quizzical.  How do people find out about changes in the search engine?  Does DT notify the upper echelon contributors?  Is there a special forum where such things are shared?  I'm just being curious.  I have no idea how DT search engine works.  Therefore, I have no idea what to change to take advantage of searchs.  But, I sure would like to get educated.   ;D

I made a couple of posts recently on the DT forum about these issues and they were confirmed by Achilles. All info is in the open, you just have to find/read it. It's quite obvious that site forums are tightly watched. They reflect on Google and they can be bad marketing-wise. Microstock sites are our partners, not our enemies. The same is true for iStock and Fotolia. For critical posts, there is always the MSG, but watch out, your posts are noticed.

The name of the game on DT is image level, just like on SS it is image recency. That's why it's important on DT not to compete with yourself with similar images, or worse, duplicates. If your downloads are split over images, none of them wil reach a higher image level. Images with a high level will not only earn you (a bit) more, but they will have a boost in the search engine. The consequence is that images on DT will only start to sell after months or even years, as they gather momentum with time. DT is like a wine cellar where the good stuff has to mature for a while. SS is Beaujolais Nouveau. DT is St.Ptrus Grand Cru.

The second issue is relevancy. Since all keywords have the same weight, DT (like BigStock) use repeat words from title and description to identify the most relevant keywords. I also have the feeling that irrelevant keywords in excess are not neutral, but will degrade your ranking, as DT probably dilutes the relevancy of keywords by the number of it. I've seen some very strong sellers of the top contributors with only 10-15 keywords.
« Last Edit: May 24, 2009, 21:51 by FlemishDreams »

« Reply #34 on: May 24, 2009, 21:55 »
0
Yeah, it definitely takes a lot of detective work or research to verify it.  The sites aren't talking. ;)

You can't hide a lighthouse on a hill, for those who can see  ;)
Doing searches yourself can tell a lot. Exploring the port of the best sellers of top contributors too.
« Last Edit: May 24, 2009, 21:58 by FlemishDreams »

« Reply #35 on: May 25, 2009, 02:06 »
0

The name of the game on DT is image level, just like on SS it is image recency.
I am not sure what's exactly the role of image level in search. Some weeks I see most sales are from images with low level; other weeks majority is with high level images.

Indeed that has to do not only with image level but also with buyer preferences, but it looks a bit strange when I browse "my earnings" pages
« Last Edit: May 25, 2009, 02:10 by MikLav »

« Reply #36 on: May 25, 2009, 03:03 »
0
My sales have also dropped like a stone  :( Shame I always liked Dreamstime but I am only getting a quarter of the income I was 2 months ago!

Xalanx

« Reply #37 on: May 25, 2009, 04:41 »
0
I am not sure what's exactly the role of image level in search. Some weeks I see most sales are from images with low level; other weeks majority is with high level images.

maybe it's FUBAR ;D

« Reply #38 on: May 25, 2009, 10:57 »
0
Ok then, what does "image level" mean?  Just the view count, or some number derrived from views, downloads, portfolio size, longitude, cosmic background radiation...?

Whatever they've done, it killed my sales too.   I'm just been in this a few months and have about 80 images online. I was making way more sales back when I had just 30. 





lisafx

« Reply #39 on: May 25, 2009, 11:06 »
0
Ok then, what does "image level" mean?  Just the view count, or some number derrived from views, downloads, portfolio size, longitude, cosmic background radiation...?




Image level is derived from number of sales:
http://www.dreamstime.com/level.php?lvlimag=3#3

« Reply #40 on: May 25, 2009, 11:33 »
0
"Them that's got shall get
Them that's not shall lose
So the Bible said and it still is news"

 - Billy Holiday/Arthur Herzog, 1939


I'm no mathematician but it isn't hard to see that as the archives get bigger, new images will have less and less of a chance, unless there's a ratings boost for new images.  And it seems that on one site after another, the boost for new images is going away.   
« Last Edit: May 25, 2009, 11:50 by stockastic »

« Reply #41 on: May 25, 2009, 18:11 »
0
I don't see a significant reduction of sales for images that increase levels. My level 4 image, exclusive to DT, sold for US$7.50 as XS.  There are many good and cheaper "hot sun" illustrations, but the buyer picked mine (and maybe others).

I suppose that in many occasions, however, the level may be the selection criteria for a buyer, if there are many images that suit his needs.

« Reply #42 on: May 25, 2009, 19:41 »
0
I'm no mathematician but it isn't hard to see that as the archives get bigger, new images will have less and less of a chance, unless there's a ratings boost for new images.  And it seems that on one site after another, the boost for new images is going away.

Yes I mention your Mathew effect often. I'm watching the DT search results closely and since March, they keep changing it all the time. They are obviously trying to find a compromise between downloads as a boost in search (favoring older pictures), and relevancy (favoring also newer pictures).
I also noticed that the only "newer" image (uploaded in 2009) I sold on DT has the search terms n/a - which means it was found by buyers wandering off in my port and reverting to visual search.

« Reply #43 on: May 25, 2009, 19:53 »
0
I don't see a significant reduction of sales for images that increase levels.

I had a quick look at my sales. In the last 20, there are 11 images with the lowest level, all others have a higher level. What are others' percentages?

« Reply #44 on: May 25, 2009, 20:17 »
0
My last 20 sales
15 subs
others 1 - 6 credits.

« Reply #45 on: May 25, 2009, 20:21 »
0
15 subs - others 1 - 6 credits.

My question was about image level, but I had 10/20 subs.

« Reply #46 on: May 25, 2009, 20:29 »
0
Duh...  14-1's 5-2's and 1-3. 

« Reply #47 on: May 25, 2009, 21:01 »
0
My wife works in a large medical office and sees a lot of glossy promotional, informational and insurance literature.  She told me she recently started noticing photos of the same model showing up all over the place.  (I did not bother to research this and try to find the source).  For me this was a concrete example of how the Mathew effect (thanks Flemish) in these searches causes buyers to end up with the same images.   Is that what they really want?


« Reply #48 on: May 25, 2009, 22:37 »
0
I had a quick look at my sales. In the last 20, there are 11 images with the lowest level, all others have a higher level. What are others' percentages?

7 L1, 5 L2, 7 L3, 1 L4

RacePhoto

« Reply #49 on: May 25, 2009, 23:41 »
0
I'm to the point of I don't care anymore with DT. Just waiting for August to remove everything. I was making an effort to send them lots of new photos, when I met with Atilla the reviewer who not only refused the latest uploads but everything in queue. At that point, the same type of shots, made at the same time, using the same standards went from about 100% accepted to absolutely overnight 100% refused.

This is part of the type of "to many like this" based on one note to reviewers, who can make a quick commission by just refusing all  without actually looking at the images. Don't believe me?



Oh yes, it hasn't hurt sales or revenue, which were only starting to get better based on more photos. And it's the photos that "don't sell well on our site" that are now selling.  :D Too small of a group or sample to make a fair assessment, but the spike for refusals, all one day, was funny.

Good news. I'm finally getting some better returns on IS and SS. Yes it's the same refused photos. You knew that already, didn't you?  ;)



 

Related Topics

  Subject / Started by Replies Last post
9 Replies
7807 Views
Last post October 11, 2006, 09:30
by Striker77s
11 Replies
7011 Views
Last post October 19, 2006, 11:59
by CJPhoto
42 Replies
17069 Views
Last post November 10, 2007, 01:50
by fotografer
3 Replies
4413 Views
Last post April 20, 2008, 12:03
by jsnover
36 Replies
10092 Views
Last post April 02, 2009, 22:52
by sgcallaway1994

Sponsors

Mega Bundle of 5,900+ Professional Lightroom Presets

Microstock Poll Results

Sponsors