MicrostockGroup Sponsors


Author Topic: Could 50 keywords be to many? Or if they are relevant its ok?  (Read 2265 times)

0 Members and 1 Guest are viewing this topic.

« Reply #25 on: January 07, 2020, 12:49 »
0
So, I did a little research on my top selling set of photos.  There's about 12 main photos in that set that sell and they are all closely subject-related. 

Shutterstock only gives you the percentage of keywords used over all the downloads, so you don't really know how many times each word was used.  And they don't always give you any of the keywords, which is a huge inconsistency.  But I was lucky that all 12 photos did have key words and percentages listed.

Dreamstime also doesn't always give you the keywords used to find your photo, but I went through 20 pages to get the keywords to the photos.

Of those photos, there were 46 different keywords used to find them.  25 of those words were used more frequently (but it's hard to tell how much more due to the lack of info given).

Since I was just about the upload a few more photos to that set, I just checked the keywords and I definitely see where I want to switch up some of the words.  I'm still going with 50, but just not the same 50. :)

46 different keywords used to find 12 related photos on DT? Do those mean sales or views? What are the # of different words on the same photos on SS?

No, 46 different keywords used to find them between both sites.  And this is all in relation to sales.  The 25 were from SS...because they just lump the stats all together.  The remaining 21 words were from DT.  Obviously, DT also had some of the 25, but I'm not spending any more time on it.  I found what I wanted to know.


« Reply #26 on: January 07, 2020, 13:05 »
0
So, I did a little research on my top selling set of photos.  There's about 12 main photos in that set that sell and they are all closely subject-related. 

Shutterstock only gives you the percentage of keywords used over all the downloads, so you don't really know how many times each word was used.  And they don't always give you any of the keywords, which is a huge inconsistency.  But I was lucky that all 12 photos did have key words and percentages listed.

Dreamstime also doesn't always give you the keywords used to find your photo, but I went through 20 pages to get the keywords to the photos.

Of those photos, there were 46 different keywords used to find them.  25 of those words were used more frequently (but it's hard to tell how much more due to the lack of info given).

Since I was just about the upload a few more photos to that set, I just checked the keywords and I definitely see where I want to switch up some of the words.  I'm still going with 50, but just not the same 50. :)

46 different keywords used to find 12 related photos on DT? Do those mean sales or views? What are the # of different words on the same photos on SS?

True it's frustrating when SS doesn't show words. I think it's the same on Alamy. Sometimes people aren't using words, they are searching categories? Logically how could someone find an image, without entering keywords? Well, many ways it appears, because they do.

Here's a tip on why so many different words might be used on DT and why some of mine were found using words that aren't in the keywords or title.



50 is too many, unless the subject is complex and has many, many, words needed to describe it. Most images can be perfectly defined with 25 or less words.

True though, placement does = sales if the images are good enough. Question I ask though is, if I'm first in a search that no one uses, how does that translate to sales and money? I could have Extracentrifical Force in my images and if no one searches the word, and I'm first on a page that only has my images, that will not translate to a sale.

I can be first on a page for something that isn't the main subject or concept of my image, just because I added words. And I won't ever get a sale, because the buyer is looking for images that represent what they are looking for, not something else.

What I think people should ask themselves, and I do, is "if I was a buyer, what words would I use to find this image?" not "do I have 50 words", that are half, not relevant or describing what's actually the subject and concept of the image.

Maybe that's just me, but I think buyers want to see what they are looking for, not something else, and if someone uses the most logical words, that would explain what most buyers will also use, not some convoluted logic that obscure and irrelevant words, will make more sales. More words does not = more sales.

ps do not use 50 words on Adobe, this has been covered, seems like that's being neglected. So I'll say, 50 words is too many, while 49 might not be. But I'll never advocate having 49 words, just to have as many as possible, because it's 49.

(clipped)

I also thought that when there are 50 keywords all keywords are flat vs 49 keywords or less, then keywords are weighted. Seems not smart. A/B testing would have solved this quirk. There is no need to assume someone is spamming if they have 50 keywords vs 49 if you just did A/B testing.

Anyway, these are just my uneducated thoughts.

Do other sites also assume that 50 keywords is spamming, and down rank your images?

https://www.microstockgroup.com/index.php?action=post;quote=539016;topic=34889.25

Here's the podcast for anyone who missed it. I need to watch the entire cast for how to increase sales.  https://www.crowdcast.io/e/keywords-count-effective/register

What I found from my research yesterday is that what *I* thought were important key words weren't ever searched for in those photos.  And words that I didn't even think would be important were being searched for. 

And we will just have to agree to disagree on the number of keywords.  It's working for me.  My photos are selling.  And the photos I mentioned that are in top spots in search results are some of my best sellers.

Uncle Pete

  • Great Place by a Great Lake - My Home Port
« Reply #27 on: January 07, 2020, 15:22 »
+1

Here's a tip on why so many different words might be used on DT and why some of mine were found using words that aren't in the keywords or title.



ps do not use 50 words on Adobe, this has been covered, seems like that's being neglected. Adobe indexing manager explains why we don't want 50 keywords.

Here's the podcast for anyone who missed it. I need to watch the entire cast for how to increase sales.  https://www.crowdcast.io/e/keywords-count-effective/register

« Last Edit: January 08, 2020, 07:11 by Uncle Pete »

« Reply #28 on: January 07, 2020, 15:33 »
+2
snip.... If DT is going to tell me my sales came from words that aren't in my title or description or keywords, how did someone find them? I'd like to know that trick.

I think they search with some keywords, then go down the rabbit hole of clicking on portfolio or similar images or others with this model. It might not take very long to get to a completely different image with really no obvious relation to the keywords listed in the search (and I definitely have some reported that don't make any sense to have shown up with a search of those words).

« Reply #29 on: January 07, 2020, 17:48 »
0

Did you ever try less words to see if your sales would be even better, as an A/B test?


Since there is no way to really have a true dependent variable, there is no way to accurately test this.  I could post the same pic on different sites, but even with the exact key words, the sales are different.  And I could post a slightly different composition, but again, with the same key words, one might sell better than the other based on the needs of the customer and the random algorithms of the site. 

In the set I mentioned, some of the shots are older.  And some of them don't have 50 words.  I checked one shot that was more popular back in the day and it only has 24.  But a photo that is only slightly older has 44 keywords and it sold better.  The most popular one in the set is newer and has 50 keywords.

I think the most important factor is the photo itself.  If it's a photo people want, they'll buy it.  However you want to keyword your photos is up to you.  I'm sticking with what is working for me.

« Reply #30 on: January 07, 2020, 19:44 »
0

Did you ever try less words to see if your sales would be even better, as an A/B test?


Since there is no way to really have a true dependent variable, there is no way to accurately test this.  I could post the same pic on different sites, but even with the exact key words, the sales are different.  And I could post a slightly different composition, but again, with the same key words, one might sell better than the other based on the needs of the customer and the random algorithms of the site. 

In the set I mentioned, some of the shots are older.  And some of them don't have 50 words.  I checked one shot that was more popular back in the day and it only has 24.  But a photo that is only slightly older has 44 keywords and it sold better.  The most popular one in the set is newer and has 50 keywords.

I think the most important factor is the photo itself.  If it's a photo people want, they'll buy it.  However you want to keyword your photos is up to you.  I'm sticking with what is working for me.

The most important factor is the photo itself, if not viewed no sale. Did you watch the AdobeStock video?

« Reply #31 on: January 07, 2020, 20:35 »
0

The most important factor is the photo itself, if not viewed no sale. Did you watch the AdobeStock video?

Yup.  And some of my top selling photos on Adobe have 50 keywords.

« Reply #32 on: January 11, 2020, 22:48 »
0
Years ago before he went private I usually noticed Yuri always maximized his keywords to 50 in still images. I tried to do the same for years. But, now-a-days I have tried to limit the number of keywords I use to 25 to make them more exact.
Besides keywords, I do make sure I use as much of the title and description characters as possible. I've found up to 64 characters when they like a title and up to 200 characters for description seems to work best for the sites I submit to, although some have more or less space. I usually split up the description into 2 sentences.
When I looked up an isolated tomato on Yuri's new site the number of keywords for one isolated tomato was 18. So, maybe they have found that less is better?

Which sites still read title? Because Adobe reads that as the description I've started to remove that data. Maybe I shouldn't do that?

Right descriptions can be just as important for searches, especially the part that search engines pick up.

What each site I contribute to seems to publish of my IPTC file info of Title, Descriptions and Keywords.
123RF = Descriptions and some Keywords
Adobe Stock = Title
Alamy = Description
BigStockPhoto = Description
CanStockPhoto = Title Description, Keywords
ClipDealer = Title, (Media Info) Description, Keywords
Crestock = Title, Description, Keywords
Cutcaster = Title, Description, Keywords
DepositPhotos = Title, Description, Keywords
Dreamstime = Title, Description, Keywords
GraphicLeftovers = Title
IStockPhotos = Title, Description, Keywords
MostPhotos = Title, Description
Panthermedia = Title, Description
Pond5 = Title, Description
Shutterstock = Description and related keywords
YAYImages = Title, Description, Keywords

« Reply #33 on: January 12, 2020, 00:40 »
0
Years ago before he went private I usually noticed Yuri always maximized his keywords to 50 in still images. I tried to do the same for years. But, now-a-days I have tried to limit the number of keywords I use to 25 to make them more exact.
Besides keywords, I do make sure I use as much of the title and description characters as possible. I've found up to 64 characters when they like a title and up to 200 characters for description seems to work best for the sites I submit to, although some have more or less space. I usually split up the description into 2 sentences.
When I looked up an isolated tomato on Yuri's new site the number of keywords for one isolated tomato was 18. So, maybe they have found that less is better?

Which sites still read title? Because Adobe reads that as the description I've started to remove that data. Maybe I shouldn't do that?

Right descriptions can be just as important for searches, especially the part that search engines pick up.

What each site I contribute to seems to publish of my IPTC file info of Title, Descriptions and Keywords.
123RF = Descriptions and some Keywords
Adobe Stock = Title
Alamy = Description
BigStockPhoto = Description
CanStockPhoto = Title Description, Keywords
ClipDealer = Title, (Media Info) Description, Keywords
Crestock = Title, Description, Keywords
Cutcaster = Title, Description, Keywords
DepositPhotos = Title, Description, Keywords
Dreamstime = Title, Description, Keywords
GraphicLeftovers = Title
IStockPhotos = Title, Description, Keywords
MostPhotos = Title, Description
Panthermedia = Title, Description
Pond5 = Title, Description
Shutterstock = Description and related keywords
YAYImages = Title, Description, Keywords

What software do you use to enter your iptc info?

« Reply #34 on: January 12, 2020, 10:06 »
0
Adobe Photoshop Elements 14 Editor, File, File Info

Uncle Pete

  • Great Place by a Great Lake - My Home Port
« Reply #35 on: January 12, 2020, 11:03 »
+1
Years ago before he went private I usually noticed Yuri always maximized his keywords to 50 in still images. I tried to do the same for years. But, now-a-days I have tried to limit the number of keywords I use to 25 to make them more exact.
Besides keywords, I do make sure I use as much of the title and description characters as possible. I've found up to 64 characters when they like a title and up to 200 characters for description seems to work best for the sites I submit to, although some have more or less space. I usually split up the description into 2 sentences.
When I looked up an isolated tomato on Yuri's new site the number of keywords for one isolated tomato was 18. So, maybe they have found that less is better?

Which sites still read title? Because Adobe reads that as the description I've started to remove that data. Maybe I shouldn't do that?

Right descriptions can be just as important for searches, especially the part that search engines pick up.

What each site I contribute to seems to publish of my IPTC file info of Title, Descriptions and Keywords.
123RF = Descriptions and some Keywords
Adobe Stock = Title
Alamy = Description

BigStockPhoto = Description
CanStockPhoto = Title Description, Keywords
ClipDealer = Title, (Media Info) Description, Keywords
Crestock = Title, Description, Keywords
Cutcaster = Title, Description, Keywords
DepositPhotos = Title, Description, Keywords
Dreamstime = Title, Description, Keywords
GraphicLeftovers = Title
IStockPhotos = Title, Description, Keywords
MostPhotos = Title, Description
Panthermedia = Title, Description
Pond5 = Title, Description
Shutterstock = Description and related keywords
YAYImages = Title, Description, Keywords

Thanks for the survey, just wondered because Adobe reads Title and not description, but SS reads Description, not title. If there is a title AS uses that, but if there's no title AS reads Description. I've been removing title from my files, before I upload to AS, and many I've never added them back.

Alternative, make a folder for Adobe, make copies, remove titles, upload and delete the files. Or make the title and description identical in all images. Which defeats the whole idea of having different fields.  :( Too bad there are so many difference in how each agency handles the metadata.

By the way, I don't know if you did that intentionally, but Adobe does read Keywords and so does Alamy. You don't have that on your list? I use software that writes all the metadata, IPTC and EXIF in XMP format.

I try to put them in the order I want them for Adobe, because that doesn't matter for SS. And Alamy, I pick the Superwords after the fact. Not sure if keyword order matters?

« Reply #36 on: January 12, 2020, 11:35 »
0
.
« Last Edit: January 12, 2020, 12:51 by Microstock Posts »

« Reply #37 on: January 12, 2020, 12:39 »
0
Sorry, I made a couple of mistakes. AdobeStock does show keywords publicly is what I am talking about, what the sites show publicly of your meta data. Alamy does show description and keywords.
What each site I contribute to seems to publish publicly of my IPTC file info of Title, Descriptions and Keywords.
CORRECTED LIST (added after next reply from ShadySue : for commercial / creative files, I did not check editorial)
123RF = Descriptions and some Keywords
Adobe Stock = Title and Keywords
Alamy = Description and Keywords
BigStockPhoto = Description
CanStockPhoto = Title Description, Keywords
ClipDealer = Title, (Media Info) Description, Keywords
Crestock = Title, Description, Keywords
Cutcaster = Title, Description, Keywords
DepositPhotos = Title, Description, Keywords
Dreamstime = Title, Description, Keywords
GraphicLeftovers = Title
IStockPhotos = Title, Description, Keywords
MostPhotos = Title, Description
PantherMedia = Title, Description
Pond5 = Title, Description
Shutterstock = Description and Keywords
YAYImages = Title, Description, Keywords
« Last Edit: January 12, 2020, 17:08 by WaterView »

ShadySue

« Reply #38 on: January 12, 2020, 13:42 »
0
IStockPhotos = Title, Description, Keywords
It's not that simple.
Description has always shown on editorial files.
For 'creative' files, for some bizarre reason, they decided not to show description on the file pages, maybe around the time of ESP. Many people complained on their forum and other official channels at the time, and no explanation or reason was given (unless I missed it, but if so, so did many other people).

Late last year, someone announced on their forum that descriptions were showing again. I'm not sure that we got any official word about this, but maybe I misssed it. But in fact, on some files the description shows as written, sometimes it doesn't show at all, and sometimes it's severely truncated: a number of mine just have the first word, which is usually 'a' or 'the'. It may be (?) that descriptions are showing correctly on newer files, but it's sketchy on older files; but I haven't the time or inclination to check.

« Reply #39 on: January 12, 2020, 17:19 »
0
Hi ShadySue, I went back to my oldest images at istock from 2014 and did see Title, Description and Keywords listed. I did not check any one else's images.

ShadySue

« Reply #40 on: January 12, 2020, 20:39 »
+2
Here are four of mine which I randomly clicked on.
The first one (2013) has the description truncated, the second one (2009) has no description at all, and for sure I put one in, the third one (2012) has been tampered with, there's no way I'd have put quotation marks around the description, and it's also been truncated, (I know how I write that sort of description, the comma would come before the scientific name, then a description of the feeder and the seeds inside), the fourth one (2009) has been truncated.
I see that the categories have been changed since upload, and some are very strange.
https://www.dropbox.com/s/lg9j0hqj3v90d78/Descriptions.jpg?dl=0

« Last Edit: January 12, 2020, 20:50 by ShadySue »

« Reply #41 on: January 12, 2020, 22:02 »
0
Hi ShadySue, So it seems the system of meta data changed about 2014 2013 by our accounts in IS. Care to share some insights about how it all matters? :~) Best, Lee


ShadySue

« Reply #42 on: January 13, 2020, 06:04 »
+2
Hi ShadySue, So it seems the system of meta data changed about 2014 2013 by our accounts in IS. Care to share some insights about how it all matters? :~) Best, Lee
1. Makes me look like an idiot
2. Makes the site look totally unprofessional - first impressions count
3. Deliberate spamming is rampant on many (all?) sites. But even when people list e.g. 10+ species or 10+ locations of a beach, most don't spam the description, so unless there is a 'genuine mistake', the buyer should be confident that what is in the description is what is shown in the image.
4. Demonstrates their tech incompetence to any who weren't already aware
5. Demonstrates management incompetence: why did they decide to do away with the descriptions in the first place ...
6. ... and not giving a reason when asked many times is a prime example of Mushroom Management.
« Last Edit: January 13, 2020, 08:51 by ShadySue »

« Reply #43 on: January 13, 2020, 10:07 »
0
Hi ShadySue, Well that was way more than I ever gave thought to. Thanks for sharing. Best to you

« Reply #44 on: January 13, 2020, 19:34 »
+1
adobe not using the description which contains more useful information than the title REQIRES more keywords to accurately describe an image

together, description & large # of keywords maximizes chances of being found; compared with title only and small # of keywords since we're getting very different TYPES of searches from buyers:

  • when search is tightly focused --> "isolated sliced red tomato", "tomato pile in market", "tomato from Oregon", "tomato in garden"
  • when search is more generic --> "tasty food", "healthy vegetable"
  • when search is between these extremes --> "fresh tomato"   


« Reply #45 on: January 13, 2020, 20:36 »
+2
But part of the key is where you end up in the search for say "tasty tomato"  or maybe even more common by a whole lot just "tomato". If you are pushed back even a tiny amount from say page 1 to page 4 (out of 100's) that means you won't make many sales.

A much more interesting question to me is if it would be worth going for top placement in a less common search and just admit you will never show up near the top of a really generic search.

Of course it probably isn't worth it to try to be too clever about this since you would have to do it differently for each site and then they will go and change the search in some way which makes your attempts useless. I say put in the obvious keywords with a good description and in the long run that should be the best. Especially over more than one site and over time.


 

Related Topics

  Subject / Started by Replies Last post
23 Replies
5711 Views
Last post June 30, 2011, 15:07
by jcpjr
14 Replies
5224 Views
Last post March 24, 2014, 10:03
by Uncle Pete
15 Replies
5546 Views
Last post February 23, 2015, 17:11
by hatman12
47 Replies
9223 Views
Last post April 17, 2018, 07:39
by Uncle Pete
9 Replies
2612 Views
Last post October 27, 2018, 15:53
by Uncle Pete

Sponsors

Microstock Poll Results