How to improve discoverability of a photograph

[efk11]

29 minutes ago, spacecadet said:

From AIM instruction manual

 

"The discoverability bar is not in any way 'reading' or 'scoring' the quality of your metadata, it simply increases with the volume of searchable information you enter".

 

Is that authoritative enough for you?

Tone!

 

The title of this topic is "How to improve discoverability of a photograph"

 

Some members commenting here are people who have good sales which suggests good ranking and good visibility and good quality images etc.

 

I on the other hand, am looking at this as someone who had 1,500 photos with no visibility and no sales! Virtually Zilch.

Now granted perhaps all my photos are shit and this is the reason but I chose to think otherwise and hand on heart I have moved photos from being undiscoverable in a few hundred pages to the top few pages many times and continue to do so daily and I had 3 sales last month.

So I am just adding my tuppence worth to help those starting off like me trying to get a leg up and also those who feel frustrated with lack of control.

I could go to the trouble of posting my stats for the last year which now are quite dramatic but its time to put up the tree!

 

Elaine

[spacecadet]

I don't know what you mean by "tone", sorry.

We've been specifically told that dIscoverability per se doesn't impact on search. That's what the quote says to me.

The reason your images are appearing higher in searches is that you have improved the keywording. It's a matter of quality, not, as you appear to insist, quantity.

It would be wrong to let your supposition go unchallenged when it doesn't accord with the facts as reported by experienced contributors who have found discoverability per se to be unimportant as Alamy has confirmed.

You have also stated that the caption is the most important factor in search. This is not correct. It's relevant, but important keywords in the caption should be repeated as supertags. This has also been verified by contributors and confirmed by Alamy.

Edited December 9, 2017 by spacecadet

[Reimar]

3 hours ago, spacecadet said:

 

We've been specifically told that dIscoverability per se doesn't impact on search. That's what the quote says to me.

 

 

As I've suggested before, the quote does not say that to me.  Of course Alamy don't have any type of AI to determine if metadata is of low or high quality.  The discoverability bar increases with volume of data (good or bad), that's clear.  What Alamy doesn't come out and say directly is that discoverability has no impact on search results.  And they probably never will since anything to do with the search algorithm is top secret.  We're left to our own devises and time wasting tests to get clues.

[efk11]

3 hours ago, spacecadet said:

I don't know what you mean by "tone", sorry.

We've been specifically told that dIscoverability per se doesn't impact on search. That's what the quote says to me.

The reason your images are appearing higher in searches is that you have improved the keywording. It's a matter of quality, not, as you appear to insist, quantity.

It would be wrong to let your supposition go unchallenged when it doesn't accord with the facts as reported by experienced contributors who have found discoverability per se to be unimportant as Alamy has confirmed.

You have also stated that the caption is the most important factor in search. This is not correct. It's relevant, but important keywords in the caption should be repeated as supertags. This has also been verified by contributors and confirmed by Alamy.

 

My opinions are just that ..my opinions, based on many hours of testing. I am not insisting anything. Im hardly one to talk with my 8 sales!!! Perhaps I am not an experienced contributor but that doesn't mean I shouldn't give me tuppence worth. I am a seller on teacherspayteachers.com and tes.com and also a software engineer having written these type of algorithms for many years so I certainly have some insight to offer. (whats written in the code, and whats written in the manual are often at odds!)

 

And I have always been one to gracefully question what we have been told! ;-) ..so whilst I accept this as what most believe and are experiencing,

From AIM instruction manual

 

"The discoverability bar is not in any way 'reading' or 'scoring' the quality of your metadata, it simply increases with the volume of searchable information you enter".

 

It is not what I am seeing in my effort to improve my discoverability. Whilst I agree quality is of course paramount, at least to photographers and people with a keen eye or looking for a particular concept or book cover,  I would believe there are plenty of buyers who are not willing to scroll past page 3 in this fast moving society of 2017 and so, discoverability trumps quality in my opinion at least!   

[Bill Brooks]

I think the best thing to do is to try to get into the green by using the largest number of the most relevant keywords possible.

 

I first use the thesaurus and then wikipedia. If I shoot the same images frequently, then I input multiple keywords into the keyword panel in Adobe Bridge. One time entry into the Bridge keyword panel, so thereafter one click gives me a series of related keywords.

 

One click, when appropriate, gets: decoration; design; motif; natural pattern; natural patterns; ornamentation; pattern; pattern in nature; patterned; patterns; patterns in nature; texture

 

One click: abandoned; ironic; man-altered landscape; manufactured landscape; obsolete; post-industrial; postindustrial; suburban; topographics; urban

 

One click: Bob Hunter Memorial Park; Canada; Little Rouge Creek; North America; Ontario; park; Rouge National Park; Rouge National Urban Park; Rouge Park; Rouge Valley; Toronto; urban wilderness

 

One click: concept; conceptual; conceptualism; idea; symbol; symbolic

 

Look up “Canada Thistle” on Wickipedia and you get: Cirsium arvense; creeping thistle; Canada thistle, Canadian thistle, lettuce from hell thistle, California thistle, corn thistle, cursed thistle, field thistle, green thistle, hard thistle, perennial thistle, prickly thistle, small-flowered thistle, way thistle; stinger-needles, eudicot, weed, injurious weed, noxious weed, wildflower, flower, perennial, purple, thistle. Alamy is an international agency so local names are important.

 

The keyword panel in Bridge makes the process fast and foolproof for subjects you do often.

 

I also have my entire collection keyworded on a hard disc. I can search for old similar images by keyword. I then copy and paste the keywords and caption from the old similar shot in the collection, to the new image I am keywording.

 

All the keywords in this post look like a lot of work. But they were not. A few clicks in the keyword panel in Bridge, and then a trip to cut and paste from my archives for a Canada Thistle, that I had keyworded before.

 

So it is not all that hard to get 50 keywords, cover all bases, and still not use a lot of misleading keywords.

Do it well only once, use many times.

[Reimar]

Bill, photoshop used to have a nice buffer that retained keywords, titles etc., that I could import to the field in a new image with one click.  CC doesn't have that anymore, and I miss it.

 

I would like to find this function in Bridge, although so far the text in Bridge is so small on my 4K monitor it may not be usable.

When I am in filmstrip mode, I see a long list of keywords in alphabetical order on the left that appears to be all the keywords in the folder.  Pressing Ctrl - I just gets me the same functionality as photoshop.  Where is this "keyword panel" you speak of, and how does it work?

 

As you can see I don't use Bridge much except to find images in my collection.

Thanks.

[Betty LaRue]

This thread is rather like comments of believing my work on legacy images have no bearing on my increased sales of older images, and increased sales in general. Many here state they’ve done no work on theirs and are doing fine, thankyouverymuchBetty.

I beg to differ. 

And had you done the work how can you tell that your zooms/sales would not have directly increased? You can’t. You can’t prove a negative.

I’m sure everyone whose port is too large to do the work, and those who choose not to will keep on chugging right along making sales.

I chose to do the work, and I am seeing good results, thankyouverymuchforum! 

Betty

And I believe what Elaine is seeing from her work!

Edited December 9, 2017 by Betty LaRue
Add text

[John Mitchell]

Since the now-infamous "discoverability" bar arrived on the scene, I find myself going back and adding more keywords to images. Have to admit that I'm often too economical with my keywords. Today, I was reworking some old images of cacti and discovered that I had missed all kinds of alternate names for one common type of cactus. However, I still think that it's easy to overdue things. For instance, I'm not sure that a buyer is going to use a vague term like "conceptualism" (sorry, Bill ) as a search term. But I guess you never know...

[Reimar]

1 hour ago, Reimar said:

Bill, photoshop used to have a nice buffer that retained keywords, titles etc., that I could import to the field in a new image with one click.  CC doesn't have that anymore, and I miss it.

 

I would like to find this function in Bridge, although so far the text in Bridge is so small on my 4K monitor it may not be usable.

When I am in filmstrip mode, I see a long list of keywords in alphabetical order on the left that appears to be all the keywords in the folder.  Pressing Ctrl - I just gets me the same functionality as photoshop.  Where is this "keyword panel" you speak of, and how does it work?

 

As you can see I don't use Bridge much except to find images in my collection.

Thanks.

OK, never mind.  I think I got it.  I can create named metadata templates in Bridge and these xmp files will appear for import in Photoshop (where I can actually read the text).

Cheers.

[Bill Brooks]

The keyword tree is much more effective, flexible, and faster than the metadata templates. Think of it as an inverted tree with the trunk on top. The top Keyword could be “North America” then it branches into 3 keywords “Mexico” “USA” Canada” Then “Canada” would branch into “Ontario” “Quebec” etc. Click on “Ontario” and you get Canada; North America; Ontario. The software ignores “Quebec” and “USA” and “Mexico” because they are not connected to Ontario.

 

If I click on the word “Rouge River” at the bottom of the inverted tree it will select all the way up the tree to the trunk at the top “North America” It will insert all of the following keywords all at one time. Canada; North America; Ontario; park; Rouge National Park; Rouge National Urban Park; Rouge Park; Rouge River; Toronto;

If the Rouge River is not in the image but the image was taken in the park. Then move up the inverted tree a few notches and click on “Rouge Park” and get: Canada; North America; Ontario; park; Rouge National Park; Rouge National Urban Park; Rouge Park; Toronto

 

If there is still a particular keyword you do not want then unclick it on the keyword tree. It and only it, will be removed from the resulting keyword list.

 

You can also keyword any number of images all at once simply by selecting them.

 

To John’s point if the keyword is correct and you have the space you are not overdoing it. The  keyword "conceptualism" is put in the particular image keyword list automatically, once you put it on the keyword tree. Input a keyword only once on the keyword tree, and output many times to particular images at no extra work.

 

You do not have to build the keyword tree all at once. Build it as your shooting enters new areas and you have a need for it. For instance my keyword tree has no branches for “Finland” because I have never been there.

 

Reimar; You do not see the keyword tree because you have not yet built it. I think you are looking at the other keywords section that has all the keywords of the current images in the content panel.

 

As an experiment type the keyword NORTH AMERICA in the input box at the bottom of the keyword panel. Hit + symbol at panel bottom you should see NORTH AMERICA appear as a keyword in the keyword panel. Then highlight NORTH AMERICA in the keyword panel. Now type CANADA in the input box and hit L+ (subkeyword) symbol in the panel bottom and CANADA should appear as a subkeyword of NORTH AMERICA. You have now started to build a keyword tree.

 

If you are still interested but confused, pull down the HELP menu in BRIDGE. There is an entire page on how to build a keyword tree. You can also do much the same in Lightroom.

[Cryptoprocta]

See, I'm not sure that anyone's anecdotal evidence, including mine, 'proves' anything.

Over the past few years, I've noticed that for no explicable reason, I tend to have one month per year, usually one of the summer months, which has a huge spike in zooms/CTR. It doesn't seem to result in a spike of sales.

Anyway, this year it was May.

In my own circumstances, 'life happened' meaning I was unable to make/upload images from last Summer, and only re-started at the end of May. At the same time and ongoing, I went back over old files and making supertags and sorting the mistakes where keywords were merged. I did about 1/3 of my existing port, and worked on my new images.

But my zooms/CTR chart is going down, not up. (Sales are up this year, over the whole year, not just since I started adding supertags etc.).

So am I to speculate that working on my old files has had a negative effect, or is it all just coincidental? Probably the latter.

Again, I think that all 'discoverability' means is that if you insert all the relevant tags, your files will be discoverable on these tags, remembering that there is no stemming on Alamy's search. But I could be wrong.

[gvallee]

37 minutes ago, Cryptoprocta said:

See, I'm not sure that anyone's anecdotal evidence, including mine, 'proves' anything.

Over the past few years, I've noticed that for no explicable reason, I tend to have one month per year, usually one of the summer months, which has a huge spike in zooms/CTR. It doesn't seem to result in a spike of sales.

Anyway, this year it was May.

In my own circumstances, 'life happened' meaning I was unable to make/upload images from last Summer, and only re-started at the end of May. At the same time and ongoing, I went back over old files and making supertags and sorting the mistakes where keywords were merged. I did about 1/3 of my existing port, and worked on my new images.

But my zooms/CTR chart is going down, not up. (Sales are up this year, over the whole year, not just since I started adding supertags etc.).

So am I to speculate that working on my old files has had a negative effect, or is it all just coincidental? Probably the latter.

Again, I think that all 'discoverability' means is that if you insert all the relevant tags, your files will be discoverable on these tags, remembering that there is no stemming on Alamy's search. But I could be wrong.

 

CTR is only one item of the ranking algorithm. Lots of other secret ingredients go into determining ranking. Sale price was one of them (still is?).

Your position would only change after a re-ranking in the past (now I'm not sure), and it used to reflect activity of the last 100 days. So variations in daily/monthly CTR would have no immediate effect on ranking/visibility/sales.

 

Gen

[Cryptoprocta]

Ah, well price of sales will have done me in. In August, I got >20 payments for unpaid uses of a file back for a few years, all grossing <$6, so my already relatively low RPD went way down then. That was an unforeseen consequence of me insisting on chasing them up and Alamy insisting on billing them at a super low price.

[gvallee]

Just now, Cryptoprocta said:

Ah, well price of sales will have done me in. In August, I got >20 payments for unpaid uses of a file back for a few years, all grossing <$6, so my already relatively low RPD went way down then. That was an unforeseen consequence of me insisting on chasing them up and Alamy insisting on billing them at a super low price.

 

Price used to be taken into account (that's why Novel Use was excluded from ranking IIRC). These days, everything is speculation. Might just be a coincidence.

 

Gen

[Reimar]

4 hours ago, Bill Brooks said:

The keyword tree is much more effective, flexible, and faster than the metadata templates. Think of it as an inverted tree with the trunk on top. The top Keyword could be “North America” then it branches into 3 keywords “Mexico” “USA” Canada” Then “Canada” would branch into “Ontario” “Quebec” etc. Click on “Ontario” and you get Canada; North America; Ontario. The software ignores “Quebec” and “USA” and “Mexico” because they are not connected to Ontario.

 

If I click on the word “Rouge River” at the bottom of the inverted tree it will select all the way up the tree to the trunk at the top “North America” It will insert all of the following keywords all at one time. Canada; North America; Ontario; park; Rouge National Park; Rouge National Urban Park; Rouge Park; Rouge River; Toronto;

If the Rouge River is not in the image but the image was taken in the park. Then move up the inverted tree a few notches and click on “Rouge Park” and get: Canada; North America; Ontario; park; Rouge National Park; Rouge National Urban Park; Rouge Park; Toronto

 

If there is still a particular keyword you do not want then unclick it on the keyword tree. It and only it, will be removed from the resulting keyword list.

 

You can also keyword any number of images all at once simply by selecting them.

 

To John’s point if the keyword is correct and you have the space you are not overdoing it. The  keyword "conceptualism" is put in the particular image keyword list automatically, once you put it on the keyword tree. Input a keyword only once on the keyword tree, and output many times to particular images at no extra work.

 

You do not have to build the keyword tree all at once. Build it as your shooting enters new areas and you have a need for it. For instance my keyword tree has no branches for “Finland” because I have never been there.

 

Reimar; You do not see the keyword tree because you have not yet built it. I think you are looking at the other keywords section that has all the keywords of the current images in the content panel.

 

As an experiment type the keyword NORTH AMERICA in the input box at the bottom of the keyword panel. Hit + symbol at panel bottom you should see NORTH AMERICA appear as a keyword in the keyword panel. Then highlight NORTH AMERICA in the keyword panel. Now type CANADA in the input box and hit L+ (subkeyword) symbol in the panel bottom and CANADA should appear as a subkeyword of NORTH AMERICA. You have now started to build a keyword tree.

 

If you are still interested but confused, pull down the HELP menu in BRIDGE. There is an entire page on how to build a keyword tree. You can also do much the same in Lightroom.

 

Thanks Bill for your detailed explanation.  I see what you mean. I'll give it a go after my next shoot. 

My problem with Bridge is that the font size is so small, I can barely make out the spelling.

[Betty LaRue]

This has been discussed before that when one doesn’t  upload on a fairly regular basis, it seems that it negatively affects one’s placement, CTR or whatever!  

I do know that when I have not uploaded on a regular basis it has negatively affected my port’s performance. I do my best when life gets complicated to try to upload at least five images a week. 

I suggested in a recent thread that if one has an idea that life is going to interfere, a move, surgery, etc., to gather some finished images ready for uploading. Then dribble a few out each week. That only takes a few minutes, although once QCd, they do need finished. 

Even that could be made simple.

Take and process a set of 3-5 images that all will have the same tags. Select them all when preparing for sale and you only have to effectively prepare one image, because you only do the work for one.

If your “stash” has multiple sets, your time spent is very minimized. 

 

Early last year I began preparing our home before we listed it for sale. That required going through every closet, drawer, and the attic. Gave things away, donated, put things aside for a garage sale. Packed some of it away to move.

Then the garage sale, which organization took forever. I had family help there. Then I began fixing up the house. Hiring doors replaced, myself doing painting and repairs indoors and out, while caretaking and dealing with a spouse with dementia. And that, folks, nearly sent me around the bend.  Because every job I did took 4 times as long as it should have because my husband wanted to help but messed up everything he tried to do and I had to repair/repaint. And try to convince him to just leave me alone and please let me do it. Which he refused to do. Tears, and all that.

With all of that going on, I still managed to upload a few images a week. Somehow. 

House still for sale. Port is in good shape.

Betty

[Rick Lewis]

This has been a very interesting thread.  Where I have a huge problem is with the "concept" type keywords.  I continue to look at the search terms in "All of Alamy", for searches using conceptual terms but find very few.

 

I'm curious to know how the more experienced contributors treat this in keywording.  I'm more of a minimalist in nature and believe in the KISS (keep it simple stupid) way of thinking.  I guess if you knew my complete background, you would understand why.

 

I just can't get my head around thinking up 20 or so additional keywords (to get into the green), based on using one of my images in a "creative" way.  It just seems to me that will only have my images showing up in searches as totally irrelevant.  Enough of that is already happening.  :-)  

 

Rick

[spacecadet]

7 hours ago, Rick Lewis said:

This has been a very interesting thread.  Where I have a huge problem is with the "concept" type keywords.  I continue to look at the search terms in "All of Alamy", for searches using conceptual terms but find very few.

 

I'm curious to know how the more experienced contributors treat this in keywording.  I'm more of a minimalist in nature and believe in the KISS (keep it simple stupid) way of thinking.  I guess if you knew my complete background, you would understand why.

 

I just can't get my head around thinking up 20 or so additional keywords (to get into the green), based on using one of my images in a "creative" way.  It just seems to me that will only have my images showing up in searches as totally irrelevant.  Enough of that is already happening.  :-)  

 

Rick

Based on a cursory look, your tagging is a little spare but it wouldn't trouble me much. Maybe a few tags relating to secondary elements, vehicles as well as buildings, for example, but I wouldn't lose sleep over it.

How many views you get is relevant. A few years ago I deleted a lot of peripheral keywords and my views nearly halved- - something under one view per image per month is what I get now. I would regard much more as spammy.

Of course I stand by my view that discoverability per se isn't relevant to search.

[Cryptoprocta]

13 hours ago, Betty LaRue said:

This has been discussed before that when one doesn’t  upload on a fairly regular basis, it seems that it negatively affects one’s placement, CTR or whatever!  

I do know that when I have not uploaded on a regular basis it has negatively affected my port’s performance. I do my best when life gets complicated to try to upload at least five images a week. 

I suggested in a recent thread that if one has an idea that life is going to interfere, a move, surgery, etc., to gather some finished images ready for uploading. Then dribble a few out each week. That only takes a few minutes, although once QCd, they do need finished. 

 

...

 

Seems like an odd policy. Given that I didn't know that my husband was going to be diagnosed with a serious illness, die nine weeks later, then my Mum would be in hospital for three months and be left virtually chairbound - a round trip of about five hours and with no access online - I hadn't withheld files to drip up.

It's counter-intuitive that  a large number files could serve us better hanging around on our hard-drives 'in case of the unexpected' rather than be online and onsale, but there are a few Alamy policies which defy my attempts to understand them.

 

Thanks for the info, though.

[wiskerke]

8 minutes ago, Cryptoprocta said:

Seems like an odd policy. Given that I didn't know that my husband was going to be diagnosed with a serious illness, die nine weeks later, then my Mum would be in hospital for three months and be left virtually chairbound - a round trip of about five hours and with no access online - I hadn't withheld files to drip up.

It's counter-intuitive that  a large number files could serve us better hanging around on our hard-drives 'in case of the unexpected' rather than be online and onsale, but there are a few Alamy policies which defy my attempts to understand them.

 

Thanks for the info, though.

 

That's a sad story.

The other story about uploading generating sales is just magical thinking.

It's just the quality not the quantity nor the frequency.

However if your sales depend on the New tab, I can see some merit in it, but then it would mainly be the newer images selling.

I had a year like yours from mid 2015 till the end of 2016 and my sales were at an all time high.

 

wim

[Betty LaRue]

2 hours ago, Cryptoprocta said:

Seems like an odd policy. Given that I didn't know that my husband was going to be diagnosed with a serious illness, die nine weeks later, then my Mum would be in hospital for three months and be left virtually chairbound - a round trip of about five hours and with no access online - I hadn't withheld files to drip up.

It's counter-intuitive that  a large number files could serve us better hanging around on our hard-drives 'in case of the unexpected' rather than be online and onsale, but there are a few Alamy policies which defy my attempts to understand them.

 

Thanks for the info, though.

So sorry for what you are going through.

I don’t believe uploading every week influencing placement is a policy of Alamy’s.  But considering how little we understand the algorithm of searches, who knows about under-the-table influences. I do think frequent uploads combined with one’s CTR and where one’s images are normally placed may mix in a favorable soup. This soup might be tasty for some contributors, not so much for others.

Just about everything concerning searches is confusing, and anything discovered or thought by one contributor will have different results for another.

I agree having a folder of images languishing on our hard drives waiting for bad times isn’t the thing to do. My suggestion was if you foresee a time, such as a planned surgery or a move, it can be helpful. If life rears up suddenly and smacks you in the head, that’s different.

And Wim, if what you say about images being in the new section would only spark new image sales, how about “more images by this contributor “ part? I might see the newer image zoomed and multiple older images of mine showing below. If a buyer prefers one of those, then fresh uploads might influence older image sales.

And having a steady infusion of images showing in the new section can only be good. Sure can’t see it being detrimental!

But then, most of my supposedly hare-brained musings are shot down here!  <says she as she puts on her flak jacket>

Betty

Edited December 10, 2017 by Betty LaRue
Typo

[losdemas]

38 minutes ago, Betty LaRue said:

So sorry for what you are going through.

I don’t believe uploading every week influencing placement is a policy of Alamy’s.  But considering how little we understand the algorithm of searches, who knows about under-the-table influences. I do think frequent uploads combined with one’s CTR and where one’s images are normally placed may mix in a favorable soup. This soup might be tasty for some contributors, not so much for others.

Just about everything concerning searches is confusing, and anything discovered or thought by one contributor will have different results for another.

I agree having a folder of images languishing on our hard drives waiting for bad times isn’t the thing to do. My suggestion was if you foresee a time, such as a planned surgery or a move, it can be helpful. If life rears up suddenly and smacks you in the head, that’s different.

And Wim, if what you say about images being in the new section would only spark new image sales, how about “more images by this contributor “ part? I might see the newer image zoomed and multiple older images of mine showing below. If a buyer prefers one of those, then fresh uploads might influence older image sales.

And having a steady infusion of images showing in the new section can only be good. Sure can’t see it being detrimental!

But then, most of my supposedly hair-brained musings are shot down here!  <says she as she puts on her flak jacket>

Betty

 

You can borrow my tin hat, Betty!

[Betty LaRue]

33 minutes ago, losdemas said:

 

You can borrow my tin hat, Betty!

I’ll take it!

[Rick Lewis]

13 hours ago, spacecadet said:

Based on a cursory look, your tagging is a little spare but it wouldn't trouble me much. Maybe a few tags relating to secondary elements, vehicles as well as buildings, for example, but I wouldn't lose sleep over it.

How many views you get is relevant. A few years ago I deleted a lot of peripheral keywords and my views nearly halved- - something under one view per image per month is what I get now. I would regard much more as spammy.

Of course I stand by my view that discoverability per se isn't relevant to search.

 

Thank you.  I'm still learning, for sure.  I do spend a lot of hours looking at other photographers portfolios, including the keywording.  I'm trying to eliminate some of the sheer pointless places my images sometimes end up, when searches are conducted.  I think things are getting better.  I've been reworking many of the keywords for my files to try to get the most relevant terms posted.

 

I know one thing for sure.  Only the man behind the curtain, (Wizard of Oz reference), knows for certain how the search algorithm is built.  I know I certainly don't have a clue.  

 

Rick

[Inchiquin]

OK, I can now confirm definitely that in the case of a single image, discoverability has no effect on its search position. I chose an image with 25 tags, 5 of which were supertags, and orange discoverability, and searched on two of the supertags. I then added tags to give it the full complement of 50 and made 10 of them supertags, which turned it green. I waited two days (to make sure the system was updated) and searched again on the original two tags. The position was unchanged.

 

What I don't know (and I have no intention of finding out by experimentation) is whether the proportion of green images in your collection influences your Alamy rank.

 

Alan