MicrostockGroup Sponsors


Author Topic: Measures not showing all results  (Read 10398 times)

0 Members and 1 Guest are viewing this topic.

Mimages

« on: November 26, 2018, 03:15 »
0
An on-going problem for years with Alamy Measures is that the table doesn't show the last few lines of results. Usually you wouldn't even know, but on days like today when I could only find some of my 7 zooms, I know that there is more data there that's hidden. The way to see the missing data is to sort on another column. I've checked this again today in 2 different browsers, to rule out particular browser formatting issues, and on both browsers the last 9 lines of Measures are not displayed when using the default sorting on the Total Zooms column.

This isn't a major issue but is still an issue and probably very simple to fix. Alamy are aware of it but it takes a lot of reports of a bug before they'll do anything. So please comment if you see this problem too.


« Reply #1 on: November 27, 2018, 07:05 »
0
Not understanding? I have 20 lines per page.

Mimages

« Reply #2 on: November 27, 2018, 07:10 »
0
Not understanding? I have 20 lines per page.

Me too, but sometimes the last ones don't show. That's with the columns sorted by Total Zooms. Today, they are all showing but often they don't. I think it might just be when the last page ends with only a few left to display, but there is no further page available so you can't see the last few lines without changing the sorting. Today, I have 7 lines left for the last page, and I see them all. If there is something like 3 or 4 lines left, they don't show. Hope that makes sense!  :) 

ShadySue

  • There is a crack in everything
« Reply #3 on: November 27, 2018, 07:27 »
0
An on-going problem for years with Alamy Measures is that the table doesn't show the last few lines of results. Usually you wouldn't even know, but on days like today when I could only find some of my 7 zooms, I know that there is more data there that's hidden. The way to see the missing data is to sort on another column. I've checked this again today in 2 different browsers, to rule out particular browser formatting issues, and on both browsers the last 9 lines of Measures are not displayed when using the default sorting on the Total Zooms column.
My default seems to be Your Zooms (that's where it is whenever I got into my measures. I guess I must have set it up that way, but I don't remember!
Or are you talking about All of Alamy?
If I want to see my zooms, it makes more sense for me to choose 'Your images' rather than AoA. IMO - YMMV.

Mimages

« Reply #4 on: November 27, 2018, 07:31 »
0
My default seems to be Your Zooms (that's where it is whenever I got into my measures. I guess I must have set it up that way, but I don't remember!
Or are you talking about All of Alamy?
If I want to see my zooms, it makes more sense for me to choose 'Your images' rather than AoA. IMO - YMMV.

I think the default is whatever you last chose, probably stored in a cookie. I do mean "Your Images", not AofA. Can't see your own zooms in AofA.

It is a definite bug, as it's repeatable in other browsers so can't just be happening to me. Keep an eye out, or check previous day's measures and you'll likely see the same thing at some point.

ShadySue

  • There is a crack in everything
« Reply #5 on: November 27, 2018, 07:44 »
0
My default seems to be Your Zooms (that's where it is whenever I got into my measures. I guess I must have set it up that way, but I don't remember!
Or are you talking about All of Alamy?
If I want to see my zooms, it makes more sense for me to choose 'Your images' rather than AoA. IMO - YMMV.

I think the default is whatever you last chose, probably stored in a cookie. I do mean "Your Images", not AofA. Can't see your own zooms in AofA.

It is a definite bug, as it's repeatable in other browsers so can't just be happening to me. Keep an eye out, or check previous day's measures and you'll likely see the same thing at some point.
But if I'm sorting by Your Zooms, all my zooms are at the top of the page.
And I usually only look each day for 'yesterday' or Friday to Sunday on a Monday, so my zooms aren't likely to fall off the first 20. If I look for a month, the zooms cover several pages, but there are lots of pages after that with search terms where none of my images were zoomed, so I think I still see all my zooms (how would I know if not?)

I think I must be missing something of what you're saying, sorry.

Mimages

« Reply #6 on: November 27, 2018, 07:51 »
0

But if I'm sorting by Your Zooms, all my zooms are at the top of the page.
And I usually only look each day for 'yesterday' or Friday to Sunday on a Monday, so my zooms aren't likely to fall off the first 20. If I look for a month, the zooms cover several pages, but there are lots of pages after that with search terms where none of my images were zoomed, so I think I still see all my zooms (how would I know if not?)

I think I must be missing something of what you're saying, sorry.

It's ok, I might not be explaining it very well.  :)  You're right that when sorting by your own zooms, you'll see them all. What I tend to do is sort by all zooms, then go through and look at how many others were zoomed as well as mine, as that obviously gives a better clue of whether my image maybe used or not. Seeing the results in the "all zooms" order feels like a faster way to do that - Although it might be just as fast when sorting by my own zooms I suppose! The more I think about it the more I wonder why I do it the way I do.

Well it seems it's nothing for most people to be concerned about. If you want to see all your views though, or which searches had lots of other zooms but not yours (to help you do tighter keywording maybe), or you want to see all entries for any other sorting, you'll see some results missing at the end (unless it's just me). If you're only using Measures to study your own zooms then it won't be an issue.

ShadySue

  • There is a crack in everything
« Reply #7 on: November 27, 2018, 08:53 »
0
If you want to see all your views though, or which searches had lots of other zooms but not yours (to help you do tighter keywording maybe), or you want to see all entries for any other sorting, you'll see some results missing at the end (unless it's just me). If you're only using Measures to study your own zooms then it won't be an issue.
Better keywording is always a good aim, but the way their system works (and for many months now, caption is trumping keywords), a properly captioned/keyworded photo of Joe Bloggs and Jane Doe will show up in searches for Joe Doe, Jane Bloggs, Doe Bloggs, and Jane Joe (etc), so you can't overthink it. Any random word in the keywords or caption can be joined with any other, UNLESS the searcher uses quotation marks, which is extremely rare (for a reason I don't understand).

Mimages

« Reply #8 on: November 27, 2018, 09:07 »
0
Better keywording is always a good aim, but the way their system works (and for many months now, caption is trumping keywords), a properly captioned/keyworded photo of Joe Bloggs and Jane Doe will show up in searches for Joe Doe, Jane Bloggs, Doe Bloggs, and Jane Joe (etc), so you can't overthink it. Any random word in the keywords or caption can be joined with any other, UNLESS the searcher uses quotation marks, which is extremely rare (for a reason I don't understand).

I've done a lot of testing of keyword and caption search engine weighting, and it often changes. For a while now, the caption hasn't trumped supertags like it did for several months. The weighting is currently supertags, then captions, then tags. You're right that you can overthink it, but it pays off to do a thorough job.

A fairly recent change is that duplicate words as part of phrases no longer help like they used to, which is actually a big shame because that used to help my placement a lot. Having the actual phrases match what someone searches for is a very big help though. The same word in both the caption and tags, or tags and supertags, or all 3, does also help placement. Myself and others have done very comprehensive testing on all this in the past and more recently.

There's a lot of misinformation about all this posted on Alamy's forum, which is misleading to others. The facts do change though, and since the search engine was changed (the same time the new AIM was introduced) the way searches work has been a lot more fluid than it used to be. Having activity (zooms and sales) on an image will increase it's placement for the same search term that led to the zoom/sale. That's a good thing in my opinion.

There are other factors that may change placement, and Alamy may tweak things again so that some of what I've written is inaccurate. It's best to just be thorough, use phrases a lot, don't try to cheat the system, and add all relevant words to the caption. Then you can't really go wrong no matter what tweaks they make.

ShadySue

  • There is a crack in everything
« Reply #9 on: November 27, 2018, 09:29 »
0
I also have done lots of searches, and I've seen photos at the top of searches, with hundreds or more results, which have no keywords at all (old Live News images which were never keyworded), above my images which have the keywords 'supered'. So either these people have super high Alamy Ranks or captions can still trump supertags.
It certainly changes a lot, and I'm perplexed about why my ranking is so low in my main interest when I can see lots of false results on the first page of many searches (would a person really 'super' a tag which is evidently wrong, not an 'easy mistake'?). Yet in other searches of similar size, my files show up surprisingly high. So Alamy Rank must play a considerable part, and in some types of search my rank must be higher than others who feature in the search, but in my favoured subject, my rank must be low.
Also and for at least a few months, some searches have a difference between the Creative search sort and the Relevant sort (much cleaner search in Relevant, as you'd hope), but other searches have the same order for both and I don't know why that is. But then looking at my own placement in Creative sorts, I have no idea what 'Creative' means. It certainly doesn't (always?) mean 'sales rank high' as is often suggested.
« Last Edit: November 27, 2018, 09:42 by ShadySue »

Mimages

« Reply #10 on: November 27, 2018, 09:46 »
0
I also have done lots of searches, and I've seen photos at the top of searches, with hundreds or more results, which have no keywords at all (old Live News images which were never keyworded), above my images which have the keywords 'supered'. So either these people have super high Alamy Ranks or captions can still trump supertags.
It certainly changes a lot, and I'm perplexed about why my ranking is so low in my main interest when I can see lots of false results on the first page of many searches (would a person really 'super' a tag which is evidently wrong, not an 'easy mistake'?). Yet in other searches of similar size, my files show up surprisingly high.
Also and for at least a few months, some searches have a difference between the Creative search sort and the Relevant sort, but other searches have the same order for both and I don't know why that is. But then looking at my own placement in Creative sorts, I have no idea what 'Creative' means. It certainly doesn't (always?) mean 'sales rank high' as is often suggested.

The placement can certainly be confusing at times, and those un-keyworded Live News images are a pain. I personally wish Live News pics wouldn't go on sale as stock until I choose, so I can delete them without anyone finding them in searches. If I want them to be on sale, I'll keyword them. Most times though I prefer to re-process and upload as stock images. It means that I either have to delete them within 48 hours so they never go into stock, or I have to change captions in a way that they're unlikely to be found. Alamy refuse to change things like that, so it forces us to change details in a way that might be frowned upon.

This is what CR have to say about their Creative section...

"The Creative tab has images in it which are handpicked by our team. This is depending on the creative touch in the image and have value to our commercial customers".

My AlamyRank is pretty high so I mostly see images at the top of the first page of results, but occasionally not. It's often unclear what's made the difference. Quite often there is a wait (not just the daily database update) before images appear higher up. I have no idea why that is. AlamyRank is another thing that there's a lot of confusing or conflicting information on. Many contributors have suggested that there will no longer be re-ranks in the same way as there used to be, but I think there probably will be. It's a long time coming though since the previous one.

ShadySue

  • There is a crack in everything
« Reply #11 on: November 27, 2018, 09:55 »
0
This is what CR have to say about their Creative section...
"The Creative tab has images in it which are handpicked by our team. This is depending on the creative touch in the image and have value to our commercial customers".
So I thought ah, maybe it's just commerically-available images which show a difference between creative and relevant.
But no, I just searched Belfast Murals, and there is currently a small difference. I don't notice any 'creative touch' particularly at the top of the creative search vs the relevant search. Though of course 'creative touch' is subjective.
London has more of a difference, but I wouldn't say 'creative touch' is more obvious in the high creative rank, New York and Madrid have no difference at the top of the search.
« Last Edit: November 27, 2018, 09:59 by ShadySue »

Mimages

« Reply #12 on: November 27, 2018, 10:00 »
0
So I thought ah, maybe it's just commerically-available images which show a difference between creative and relevant.
But no, I just searched Belfast Murals, and there is currently a small difference. I don't notice any 'creative touch' particularly at the top of the creative search vs the relevant search. Though of course 'creative touch' is subjective.

Some time ago I asked Alamy to consider some of my more creative images for their creative section, but I don't think they agreed they were suitable (it's hard to know by looking at search results). I haven't looked at the difference between creative and relevant for a while now, but I'll do a few searches soon to test that out.

ShadySue

  • There is a crack in everything
« Reply #13 on: November 28, 2018, 06:57 »
+1
Better keywording is always a good aim, but the way their system works (and for many months now, caption is trumping keywords), a properly captioned/keyworded photo of Joe Bloggs and Jane Doe will show up in searches for Joe Doe, Jane Bloggs, Doe Bloggs, and Jane Joe (etc), so you can't overthink it. Any random word in the keywords or caption can be joined with any other, UNLESS the searcher uses quotation marks, which is extremely rare (for a reason I don't understand).
Today's example: I had a search for Eva Green (don't ask, I haven't a clue).
Turns out I have a photo of a garden designed by Eva Zandman (caption only) located at Marston Green (keywords only).
 ::)
(admittedly, that was a fussy searcher, who checked 3180 files. Mine isn't on the first page. [By which I learned she's some sort of celeb.])
« Last Edit: November 28, 2018, 07:01 by ShadySue »

Uncle Pete

  • Great Place by a Great Lake - My Home Port
« Reply #14 on: November 28, 2018, 08:54 »
+1

A fairly recent change is that duplicate words as part of phrases no longer help like they used to, which is actually a big shame because that used to help my placement a lot. Having the actual phrases match what someone searches for is a very big help though. The same word in both the caption and tags, or tags and supertags, or all 3, does also help placement. Myself and others have done very comprehensive testing on all this in the past and more recently.


So spammed phrases with same words repeating has been defeated, I think I like that?  :)

Since you mention it, yes I have taken a set of photos and placed the keyword or name in all three, caption, supertag, and tags. To make what I thought the best image of that series, come first. I wasn't sure about the duplicate supertag/tags if it made any difference, because supertag itself would do that?

Meanwhile, the idea is to make my favorite show in my image results, highest and if diversity algorithm selects that way, would bring that image to the lowest pages in the search. What do you think? Trying to put my best image forward. All the rest have just the caption and supertag. This is for a specific name person, or very specific subject by the way.

Mimages

« Reply #15 on: November 28, 2018, 09:27 »
+1
So spammed phrases with same words repeating has been defeated, I think I like that?  :)

Since you mention it, yes I have taken a set of photos and placed the keyword or name in all three, caption, supertag, and tags. To make what I thought the best image of that series, come first. I wasn't sure about the duplicate supertag/tags if it made any difference, because supertag itself would do that?

Meanwhile, the idea is to make my favorite show in my image results, highest and if diversity algorithm selects that way, would bring that image to the lowest pages in the search. What do you think? Trying to put my best image forward. All the rest have just the caption and supertag. This is for a specific name person, or very specific subject by the way.

I agree that it's good to stop spamming, as long as using the same word in different phrases still helps  when that phrase is searched for (which is does). If using that same word in different phrases doesn't help placement if someone searches on only that one word, then I guess that's fair enough and I agree with it.

Putting the same word as both a tag and supertag is spamming and playing the system, and I don't do that UNLESS that word is part of a different phrase. Having the same single word in the caption as well as EITHER a supertag or tag is fine of course, and needed.

As for your question of promoting certain images - Well there's no clear cut answer I'm afraid. Logically, you should be able to select the displayed order for certain searches quite easily, but these days it often doesn't work in the expected way. If you make various changes to one image and compare the placement with different tagging the next day, then you get consistent predictable results. If you have more than one image with those words in though, then it's quite unpredictable. So you might have 3 images of a balloon and put "balloon" in a supertag for your favourite, and just in the caption for your second favourite, and just as a tag for your least favourite. I have found that doing that doesn't always give the results in the order you want though. No idea why!  :)

I have noticed recently that newer images with the same tags appear higher than older ones, just like they used to. That behaviour did stop for some time, but recently I noticed it happening again. Maybe that was just coincidence, I really can't be sure.

Alamy have always preferred that we concentrate on keywording accurately and let them worry about the rest. What they don't understand is that we NEED to know at least some details of how things work, and we NEED to be able to choose the order of images in some instances. I'm all for improving the search engine but not for changing it, without notice, where we see less relevant images appearing first! That used to happen a lot, and when the new search engine went live, tags and supertags worked the opposite way around (confirmed by many contributors). Alamy did fix it in the end, but never admitted it. They were frustrating times.

Uncle Pete

  • Great Place by a Great Lake - My Home Port
« Reply #16 on: November 28, 2018, 09:44 »
+1
By the way, I didn't intentionally add Supertags and Tags the same (except one image - maybe?), until I noticed that in the conversion from the three boxes, Alamy did that. I was checking keywords in an image and saw the same two and three times. If I had the same in all three boxes, they got swept into tags three times. I doubt that it made any difference and I'm not going back to edit 2,000+ images to find what Alamy did to them!  :)

Remember when the extra data was searchable and people were crying that it ruined the search. You know the people that would copy a page from an encyclopedia into their images. So Alamy stopped that. Much of my specific data was in that area, like event details and true searchable, useful information. As a result I had to edit all my photos again to move names and places into the boxes...

If you remember, back years ago, we were promised that certain designations would make words and phrases more searchable. Never happened. Then they changed which box was visible or not, even though both were searched. I'd drop names and specifics in the invisible one. So I had to go edit all my images that I cared about.

What I mean is, they revised again and I just don't care to individually edit all my photos for a third or fourth time. I'll do the ones that I care about and leave whatever their system created, as it is.

I think you have hit on something, that the search changes at Alamy, sometimes pushing newer, sometimes featuring some other sort variation. That's no surprise because part of the game is to present fresh and changing content to buyers, even when the different view is ten year old images.

My favorite is still the diversity algorithm. Everyone gets a chance to be seen, before someone else gets better exposure. Even where I have hundreds of some category, mine are not bloating the first page. Fair for everyone.



Mimages

« Reply #17 on: November 28, 2018, 09:59 »
0
By the way, I didn't intentionally add Supertags and Tags the same (except one image - maybe?), until I noticed that in the conversion from the three boxes, Alamy did that. I was checking keywords in an image and saw the same two and three times. If I had the same in all three boxes, they got swept into tags three times. I doubt that it made any difference and I'm not going back to edit 2,000+ images to find what Alamy did to them!  :)

Remember when the extra data was searchable and people were crying that it ruined the search. You know the people that would copy a page from an encyclopedia into their images. So Alamy stopped that. Much of my specific data was in that area, like event details and true searchable, useful information. As a result I had to edit all my photos again to move names and places into the boxes...

If you remember, back years ago, we were promised that certain designations would make words and phrases more searchable. Never happened. Then they changed which box was visible or not, even though both were searched. I'd drop names and specifics in the invisible one. So I had to go edit all my images that I cared about.

What I mean is, they revised again and I just don't care to individually edit all my photos for a third or fourth time. I'll do the ones that I care about and leave whatever their system created, as it is.

I think you have hit on something, that the search changes at Alamy, sometimes pushing newer, sometimes featuring some other sort variation. That's no surprise because part of the game is to present fresh and changing content to buyers, even when the different view is ten year old images.

My favorite is still the diversity algorithm. Everyone gets a chance to be seen, before someone else gets better exposure. Even where I have hundreds of some category, mine are not bloating the first page. Fair for everyone.

I had the same when the new AIM began, that many of my previous phrases were put into individual tags. I have gone back over a lot of my older images but it's such a pain and takes so long, I tend to focus on newer (and honestly far better!) images instead. Having all those duplicates did actually help placement, at least it did then, but doesn't seem to any more (probably a good thing).

Do you mean the additional info box? I had read recently that it used to be searchable, but in those days I didn't pay much attention to how it all worked. It's only in the past few years I've really taken it all very seriously and sales have rocketed. So I'm afraid I'm not knowledgeable on how things worked over about 3 years ago. It's since 2015 that I've taken it more seriously.

You're right that the diversity algorithm is a good thing, as is (I think you probably meant this too) the fact that new contributors have a reasonably high rank, so it gives them a chance to get seen. If they do a good job, they'll soon get promoted up the ranks. New users probably don't realise just how vital it is to NOT spam keywords, and to upload the best images possible. Once your rank drops, it's very hard to push it up again because your images are rarely seen, so sales don't come. Alamy is tough, but if you do your part it can pay off and be worth it, as long as you have a lot of patience at the start.


Mimages

« Reply #18 on: November 28, 2018, 10:09 »
0
Today's example: I had a search for Eva Green (don't ask, I haven't a clue).
Turns out I have a photo of a garden designed by Eva Zandman (caption only) located at Marston Green (keywords only).
 ::)
(admittedly, that was a fussy searcher, who checked 3180 files. Mine isn't on the first page. [By which I learned she's some sort of celeb.])

That is the sort of thing that happens to me quite often. It's tough when someone searches through several pages, as even if you only have a word as a tag or part of a tag or caption, and your rank is quite high, your image will appear in the first several pages. It's always a problem when putting a location in the caption, as most people suggest you do. I usually do but it can give you a lot of unwanted views. For example, I might shoot a duck and say in the caption that it was in my county of West Sussex. I NEED to put that, as sometimes you'll get people (like wildlife magazines) use the county in a search and they only want to see certain ducks from that particular county. Having it there though means that I can get up to 50 or even more views when someone is actually looking for landmarks or views of West Sussex, and they search through 10 or 20 pages.

Luckily our rank is only slightly affected by CTR, so it shouldn't drop our rank much to have those unwanted views with no zooms. I think it might be nice to have an option box to tick that says something like, "Don't show this image unless they use a particular word in the search". So in my example maybe if "duck" isn't in the customer's search, then don't display the image in results.

I've gone off the subject from what you were getting at, sorry about that, but I do understand what you're saying.

ShadySue

  • There is a crack in everything
« Reply #19 on: November 28, 2018, 10:56 »
0
It was even worse when your name was a searchable entity, as mine is: I got lots of false hits until they changed that policy.

Uncle Pete

  • Great Place by a Great Lake - My Home Port
« Reply #20 on: November 29, 2018, 09:39 »
0
Rank is middle for anyone new until they "get enough sales". That came from Alamy in an email. What is enough sales? I suspect something like 100. So actually rank, clicks, zooms and whatever people imagine, doesn't exist for new people and doesn't do anything until we "get enough sales"

That could be called a boost or just starting on level ground instead of having to fight uphill against everyone else.

But at least new people or someone thinking of joining should know that they don't start at the bottom of the heap.

Yes if I remember right, additional information which was abused, and then the same people stood up at the meeting and complained about how their search results were bad. Yeah, the same people who put that there, got it taken out, so the rest of us lost a tool.

As for the search, we can debate until the end of time. It's a search, it sees words, a bad search will give bad results. My usual example is one word or two.

But a name of a place and object or a subject, the search can only see a word. Searches are not thinking or knowing, they can only look for what the person has keyworded and only go find what the person doing the search has entered.

Now for a phrase? Yes if buyers learned to use that? At one time Alamy did not have the option to search a phrase it was still just words.

I gave up trying to play the search game and I accept that individual words are what is available for most stock sites. Even where we can add a phrase, most still read them as individual words, not looking for an exact match.

I'm one of the people who believes, only the best and most appropriate keywords are included. (note believe... I don't have a long study or factual data, just my observations) In other words, less is more when it comes to keywords. Adding 25 more words, just for the sake of a high discovery colored box, just seems all wrong. Buyers don't search for obscure terms, they search for the obvious. People should ask themselves, Wouldn't you, what would you search for if you were looking for this image?

Scientific names are important, specific terms are important, exact and accurate names and terms are best. Adding close matches, just to have more words is likely to hurt, especially if a site uses, number of matching words, from all keywords as part of the weighting? Or just like noted, words used that led to a sale, makes the image more relevant, while words that don't lead to a sale, get no increase in relativity weighting. (that means no penalty but no boost either.)

« Last Edit: November 29, 2018, 09:43 by Uncle Pete »

Mimages

« Reply #21 on: November 29, 2018, 11:21 »
0
Rank is middle for anyone new until they "get enough sales". That came from Alamy in an email. What is enough sales? I suspect something like 100. So actually rank, clicks, zooms and whatever people imagine, doesn't exist for new people and doesn't do anything until we "get enough sales"

That could be called a boost or just starting on level ground instead of having to fight uphill against everyone else.

But at least new people or someone thinking of joining should know that they don't start at the bottom of the heap.

Yes if I remember right, additional information which was abused, and then the same people stood up at the meeting and complained about how their search results were bad. Yeah, the same people who put that there, got it taken out, so the rest of us lost a tool.

As for the search, we can debate until the end of time. It's a search, it sees words, a bad search will give bad results. My usual example is one word or two.

But a name of a place and object or a subject, the search can only see a word. Searches are not thinking or knowing, they can only look for what the person has keyworded and only go find what the person doing the search has entered.

Now for a phrase? Yes if buyers learned to use that? At one time Alamy did not have the option to search a phrase it was still just words.

I gave up trying to play the search game and I accept that individual words are what is available for most stock sites. Even where we can add a phrase, most still read them as individual words, not looking for an exact match.

I'm one of the people who believes, only the best and most appropriate keywords are included. (note believe... I don't have a long study or factual data, just my observations) In other words, less is more when it comes to keywords. Adding 25 more words, just for the sake of a high discovery colored box, just seems all wrong. Buyers don't search for obscure terms, they search for the obvious. People should ask themselves, Wouldn't you, what would you search for if you were looking for this image?

Scientific names are important, specific terms are important, exact and accurate names and terms are best. Adding close matches, just to have more words is likely to hurt, especially if a site uses, number of matching words, from all keywords as part of the weighting? Or just like noted, words used that led to a sale, makes the image more relevant, while words that don't lead to a sale, get no increase in relativity weighting. (that means no penalty but no boost either.)

The "median" AlamyRank is actually very high and nowhere near the middle. If someone searches for "dog" and there are 100K results, someone with a new account or pseudo will see their top dog image far higher than at number 50K. More like on the first few pages. The massive majority of people APPEAR (I can't be certain of course) to have a very low rank, which is why I'm not too concerned about the ever increasing number of images being added, as the positions of mine barely change. Anyone starting afresh has a high chance of monthly sales after they get 1 or 2 thousand images, as long as the quality is high and they tag accurately. With just a few sales your rank goes up, you certainly don't need anything like 100. I've watched my rank increase over the years where I had nothing like 100 sales at first. How much your rank increases is anyone's guess, and it might change depending on the fees received, or might be based on a calculation of sales per number of images.

As for using less relevant words for tags, I partially agree with you but you have to get a balance. I recently had a search for "kingfisher sussex". If I hadn't added the county in the tags, mine (that was zoomed) would not have appeared. Of course that means I'll get a lot of irrelevant views too, but you can't have it all. Where that line is that you don't specify the location is something I often change my mind about, depending on what I see in Measures for views and zooms. Yesterday's Measures data shows around 300 views for just one search that included my town and county! That will probably be all my images where I've specified my actual town. I only put the town name when it's absolutely relevant though so I can't delete the town from any of those images.

Other tags such as stating the season or temperature are tricky, so you need to judge it and get a balance. I agree with you that most searches are very general and don't use fancy words at all, so I stick to relevant ones, sometimes drifting towards a little less relevant, without drifting too far (hopefully!). Concepts are also sometimes searched for any many images can be suitable to illustrate concepts.

Phrases - A buyer doesn't have to put a phrase in quotation marks. If a buyer searches for "white dog" (without quotes) and my image appears, let's say, in position number 100 and I have separate tags of "white" and "dog", then if I change the tags on that image to have a phrase tag of "white dog" instead, my will appear higher up in results by a significant amount. That's why I use phrases a lot, and I try to think how the customer thinks.

I have read that if a customer does use quotation marks, it doesn't show results with those words in single tags, but only if the words are together in a phrase. That might be wrong or I might have misunderstood, but I read it somewhere recently. I cover all cases in my tags anyway.

Yes definitely you must use scientific names for things like animals and flowers, they're searched for more than the common names are. If it's a pink rose, and I know the cultivar, I'll still add a tag of "pink rose", as some customers use general terms like that too.

Uncle Pete

  • Great Place by a Great Lake - My Home Port
« Reply #22 on: November 29, 2018, 12:33 »
0
Ah good point, even though the email said middle I'd bet you are right about median, that's an important distinction.

So you mean people are not going to search for crimson baked dish of edible fruit in a pastry crust? They will just type in Apple Pie LOL

Yeah that's my point, but since people think that adding nonsense for more words will make them more money, nothing is going to change that.

Haven't investigated phrases, but I add them in the form of proper names, just in case.  ;) Kind of like associated words and terms, which I just mocked.


 

Related Topics

  Subject / Started by Replies Last post
26 Replies
21247 Views
Last post October 28, 2008, 16:25
by RacePhoto
3 Replies
3768 Views
Last post February 19, 2012, 09:56
by Eireann
11 Replies
4185 Views
Last post August 07, 2013, 10:14
by cathyslife
10 Replies
7451 Views
Last post March 16, 2016, 19:44
by ShadySue
20 Replies
12403 Views
Last post April 26, 2022, 14:59
by wordplanet

Sponsors

Mega Bundle of 5,900+ Professional Lightroom Presets

Microstock Poll Results

Sponsors