Alamy's low CTR

[Via]

What's up with Alamy's very low overall CTR recently?

[ann]

The bigger size of rollover pop-up is likely relevant.

- Ann

[stipe]

What does it mean?

[JohnB]

It was suggested by someone (IDP ? in the old forum) that some customers where probably adding imges straight to lightboxes for future reference and therefore by-passing the zoom. I'm no longer sure that CTR is very meaningful - maybe Alamy should find a way of including more info in the stats such as addition to lightboxes or price lookups.

[Via]

JohnB

I agree with you about the question of Alamy's CTR relevancy. Many of the images I have sold were never zoomed on and many of my zooms were never sold. Seems that sales would be a better measure instead of zooms. Am I missing something here?

[incamerastock]

It was suggested by someone (IDP ? in the old forum) that some customers where probably adding imges straight to lightboxes for future reference and therefore by-passing the zoom. I'm no longer sure that CTR is very meaningful - maybe Alamy should find a way of including more info in the stats such as addition to lightboxes or price lookups.

 

 

JohnB

I agree with you about the question of Alamy's CTR relevancy. Many of the images I have sold were never zoomed on and many of my zooms were never sold. Seems that sales would be a better measure instead of zooms. Am I missing something here?

 

CTR is incredibly meaningful. This is what Alamy say, right here http://www.alamy.com/contributor/help/alamysearch-engine-explained.asp (my emphasis):

 

Alamy advises that contributors check their keywords carefully to ensure that they are both spelt correctly and are relevant to the image concerned. This will work towards gaining a high CTR score. Keywording both literally and conceptually is still advisable, but contributors should be careful that their captions and keywords are entirely relevant to the image. It is useful for contributors to put themselves in the buyer’s position to see if they would be interested in an image for a particular search term. As a simple example, if keywording an image of a cat it would be best not to add a caption saying ‘Cat which is afraid of dogs’. Each time this image is returned on a search for ‘dogs’ it will be viewed by a client and, in the majority of cases, not clicked or purchased resulting in a low CTR. This will negatively affect the AlamyRank for the Pseudonym that is associated with this image. Finally, if the plural and singular applies to your image, then you should add both. You can use AlamyMeasures to give valuable insights into keyword, CTR and search data.

 

So, it is a measure of how targeted/accurate your keywording is compared to customer's search expectations. The better your CTR, the more times your images are being found for the right keywords and not the wrong ones. CTR is just one element of Alamy Rank. Sales is obviously another!

 

The drop in Alamy's overall CTR is likely to be due to the new (post-rebrand) Search Engine defaulting to 120 image views per page and not 30 as it did before. Hence, every search is now 90 images deeper by default and because CTR is a formula of which "Views" are a part the ratio has declined for everyone.

 

J

[Ray]

 

It was suggested by someone (IDP ? in the old forum) that some customers where probably adding imges straight to lightboxes for future reference and therefore by-passing the zoom. I'm no longer sure that CTR is very meaningful - maybe Alamy should find a way of including more info in the stats such as addition to lightboxes or price lookups.

 

 

>>>JohnB

I agree with you about the question of Alamy's CTR relevancy. Many of the images I have sold were never zoomed on and many of my zooms were never sold. Seems that sales would be a better measure instead of zooms. Am I missing something here?

 

CTR is incredibly meaningful. This is what Alamy say, right here http://www.alamy.com/contributor/help/alamysearch-engine-explained.asp (my emphasis):

 

Alamy advises that contributors check their keywords carefully to ensure that they are both spelt correctly and are relevant to the image concerned. This will work towards gaining a high CTR score. Keywording both literally and conceptually is still advisable, but contributors should be careful that their captions and keywords are entirely relevant to the image. It is useful for contributors to put themselves in the buyer’s position to see if they would be interested in an image for a particular search term. As a simple example, if keywording an image of a cat it would be best not to add a caption saying ‘Cat which is afraid of dogs’. Each time this image is returned on a search for ‘dogs’ it will be viewed by a client and, in the majority of cases, not clicked or purchased resulting in a low CTR. This will negatively affect the AlamyRank for the Pseudonym that is associated with this image. Finally, if the plural and singular applies to your image, then you should add both. You can use AlamyMeasures to give valuable insights into keyword, CTR and search data.

 

So, it is a measure of how targeted/accurate your keywording is compared to customer's search expectations. The better your CTR, the more times your images are being found for the right keywords and not the wrong ones. CTR is just one element of Alamy Rank. Sales is obviously another!

 

The drop in Alamy's overall CTR is likely to be due to the new (post-rebrand) Search Engine defaulting to 120 image views per page and not 30 as it did before. Hence, every search is now 90 images deeper by default and because CTR is a formula of which "Views" are a part the ratio has declined for everyone.

 

J

 

That is all well and good in theory but they are also 'tweaking' search engine parameters supposedly cutting out false positives. Unfortunately my false positives have actually increased recently with my name and words in the caption combining to produce bizarre combinations - supposedly neither carry much if any weight in search results.  I think the whole Alamy search engine is flawed. It doesn't stem words, which every other search engine on Earth does. And it supposedly relies on syntax or sequence which is not the results I am getting. It would be better to open the alamy database up to Google. Far better. I am not impressed at all.

 

edit - I just thought I would give you an example of the absolute crappy quality of Alamy's search engine. Someone searched for Warren Buffet and got my pics of cuttlery and crockery on a restaurant buffet. My last name is Warren. That is just unbelievable and only one example. Shutterstock and iStockphoto 's searches are way better, more intuitive and don't turn up results like this. 

[incamerastock]

That is all well and good in theory but they are also 'tweaking' search engine parameters supposedly cutting out false positives. Unfortunately my false positives have actually increased recently with my name and words in the caption combining to produce bizarre combinations - supposedly neither carry much if any weight in search results.  I think the whole Alamy search engine is flawed. It doesn't stem words, which every other search engine on Earth does. And it supposedly relies on syntax or sequence which is not the results I am getting. I am not impressed at all.

 

But keywording is up to you. Entirely. Just take out the erroneous words and/or stem to suit the image? It's non-standard but so is Alamy's use of IPTC. It applies to every single contributor or agency.

 

We have to learn how to get the best out if it. All the info is there in the contributor section.

 

J

[JohnB]

By "meaningful" I meant that if the number of zooms is so low across all of Alamy that this is no longer very informative to the contributor. My own count of zooms last month was almost the lowest I can recall - just one more than my sales total if I remember correctly. My CTR dropped not because of an excess of views but because I was not getting zooms. If the average contributer is only getting 5 zooms for every 1000 views that does not give the individual contributer very much idea of whether his/her keywording is going in the right direction (I'm ignoring quality of images in this argument).

 

For whatever reason my zooms for this month are back to normal. I have already registered more than for the whole of  last month. I am less pessimistic at the moment.

[incamerastock]

My CTR dropped not because of an excess of views but because I was not getting zooms.

 

Zooms drop off becuase the returned image (a view) is not what the searcher/client wants to see. It's all part of the same formula. You can spin it this way if you like but it's like a football manager saying:

 

"We lost, not because we let in too many goals, but because the other team scored too many goals". 

 

It's the same argument spun backwards.

 

What is an "average" contributor??

 

J

[JohnB]

 

My CTR dropped not because of an excess of views but because I was not getting zooms.

 

Zooms drop off becuase the returned image (a view) is not what the searcher/client wants to see. It's all part of the same formula. You can spin it this way if you like but it's like a football manager saying:

One of us is  missing the point. You were suggesting that the all of Alamy ctr had dropped because people were viewing more images due to the default image per page  changing. I was stating that at least in my case that was not the cause of the drop - it was due to a drop in zooms rather than an increase in views. I haven't worked through the all of Alamy stats so I don't know if that is the overall case.  As I stated another contributor who had done quite a bit of analysis seemed to think that at least some of the zooms had disappeared from the stats due to customers by-passing the zoom by using lightboxes. I don't know whether that is true.

 

An  "average contributer" is an "average contributor".  The person whose CTR matches all of Alamy.

 

Last month I had 5 zooms and 4 sales - zooms only contributed about half of my ranking points. If that had been a trend the significance of zooms in terms of ranking would be greatly diminished.

[incamerastock]

One of us is  missing the point. You were suggesting that the all of Alamy ctr had dropped because people were viewing more images due to the default image per page  changing. I was stating that at least in my case that was not the cause of the drop - it was due to a drop in zooms rather than an increase in views. I haven't worked through the all of Alamy stats so I don't know if that is the overall case.  As I stated another contributor who had done quite a bit of analysis seemed to think that at least some of the zooms had disappeared from the stats due to customers by-passing the zoom by using lightboxes. I don't know whether that is true.

 

An  "average contributer" is an "average contributor".  The person who's CTR matches all of Alamy.

 

Last month I had 5 zooms and 4 sales - zooms only contributed about half of my ranking points. If that had been a trend the significance of zooms in terms of ranking would be greatly diminished.

 

Haha

 

Yep, I guess we're both coming at this from different angles John.  Views and zooms feed into the same formula so they are linked. A rise in views or a drop in zooms has the same effect on CTR. There's no doubt a mathematical term for this but that's above my pay grade.

 

Take your point about the "average" contributor having the same CTR. It's a nice target to aim to beat. If you're always on the right side of that figure you're going in the right direction.

 

We'll never know the true Rank formula. Zooms and sales play a part, I think overall $ revenue as a proportion of collection size does too (i.e. your "hit rate") but I'd hate to have to work it all out. The formula is no doubt incredibly complex but (to get back on topic) "CTR" is one measure which we as contributors can use to measure the efficiency of our collections so long may it last. I'd hate to be doing this "blind"...

 

J

[CLSI]

For those who like crunching numbers, we can actually take some of the guesswork out of this. Since I learned that it's possible to export data from All of Alamy to Excel for a week at a time, I've taken a sample week from each month (avoiding weeks with holidays), exported the data to Excel, and calculated the total UCO, Zooms, Views, and CTR for All of Alamy. Here it all is in a (poorly formatted) table:

 

Month                 UCO         Zooms         Views              CTR (%)
May 2012           32,472       24,109         5,137,774       0.47
June                   28,783       23,498         4,162,522       0.56
July                    35,722       25,937         5,356,505       0.48
August               33,221       23,955         5,000,830       0.48
September        37,044       28,553          5,532,649       0.52
October             35,374       28,777         5,319,959       0.54
November         35,328       27,069          5,185,255       0.52
December         34,420       25,476          4,864,140       0.52
January 2013    35,781       27,492          5,342,107       0.51
February           35,583       26,186          5,630,068       0.47
March               37,740       29,984          5,490,494       0.55
April                  40,863       32,271          7,623,099       0.42
May                   36,645       27,994          6,881,034       0.41
June                 42,439        31,262          8,257,224       0.38

 

Note that these just represent one week from each month, and that the CTR in this table is the CTR as defined by Alamy ([total zooms divided by total views] X 100). Oddly, the CTR that Alamy reports as the "Average CTR for Alamy last month" is calculated differently, and to compare it to your own CTR for the past month is something of an apples-to-organges comparison. I have no idea why Alamy uses this "average CTR" instead of the true CTR for reporting the previous month.

 

The table reveals some interesting trends. With the launch of the website in March, there was an increase in UCO and zooms before the big increase in views (which happened in April). This led to a spike in the CTR in March, probably as customers were checking out the new, larger comps (many customers undoubtedly zoomed images just for the sake of seeing what the new comps looked like). I have no explanation for the drop in all stats in May, but June is clearly showing an increase in everything except CTR. Note that even though the CTR is dropping, the total number of zooms is up over pre-March numbers, as is the total number of unique searches. This strongly suggests that the new website is bringing in more activity--either new customers or the same customers turning to Alamy more frequently. As others have noted, the overall drop in CTR is likely a result of the switch to 120 pictures per page as the default.

 

Thanks to the 4th of July in the U.S., we haven't had a good holiday-free week in July yet, so we can't really say whether this trend will continue. So far, though, it's looking like there's reason for optimism. I'll also go out on a limb and say that I expect the new Alamy comps will compete better with comps from other agencies, which (if true) should show up as a general improvement in our sales-to-zooms ratios. For those of you who like keeping statistics, compile your sales and zooms stats now for as far back as you can so that we'll have good numbers to compare with post-new-website stats (because of the time it takes for sales to register, I suspect it's much too soon for us to see any effect yet).

[Ray]

 

That is all well and good in theory but they are also 'tweaking' search engine parameters supposedly cutting out false positives. Unfortunately my false positives have actually increased recently with my name and words in the caption combining to produce bizarre combinations - supposedly neither carry much if any weight in search results.  I think the whole Alamy search engine is flawed. It doesn't stem words, which every other search engine on Earth does. And it supposedly relies on syntax or sequence which is not the results I am getting. I am not impressed at all.

 

But keywording is up to you. Entirely. Just take out the erroneous words and/or stem to suit the image? It's non-standard but so is Alamy's use of IPTC. It applies to every single contributor or agency.

 

We have to learn how to get the best out if it. All the info is there in the contributor section.

 

J

J ..thanks for your answer ... but yes I realise those points and that's why I gave the example. If I had put something stupid in my keywords I would just quietly fix it and not post it here in the virtual town square.

Despite a recent blog post saying they'd removed pseudonyms from basic searches my surname Warren came up in a search for Warren Buffet. I had included the keyword 'buffet' which is relevant when they are pictures of a restaurant buffet with crockery and cutlery on them.  

I don't know if you have tried Shutterstock's searches. They get very good results and the keywording is very loose by comparison to Alamy's. I don't think it is too much to ask Alamy to be as good as the competition, especially when it affects our CTR. 

[incamerastock]

J ..thanks for your answer ... but yes I realise those points and that's why I gave the example. If I had put something stupid in my keywords I would just quietly fix it and not post it here in the virtual town square.

Despite a recent blog post saying they'd removed pseudonyms from basic searches my surname Warren came up in a search for Warren Buffet. I had included the keyword 'buffet' which is relevant when they are pictures of a restaurant buffet with crockery and cutlery on them.  

I don't know if you have tried Shutterstock's searches. They get very good results and the keywording is very loose by comparison to Alamy's. I don't think it is too much to ask Alamy to be as good as the competition, especially when it affects our CTR. 

 

Ray, that's really strange. Alamy have turned off the pseudonym in searches now so if you've seen this happen it must be a report from before the change (a couple of weeks ago) it won't happen now *unless* you accidentally have 'Warren" as another keyword somewhere in a searchable field.

 

I just tried to see which image it was by doing an Advance search on your name (as Contributor) and "buffet" as a keyword but it produces zero hits. Have you re-keyworded it?

 

J

[Ray]

 

J ..thanks for your answer ... but yes I realise those points and that's why I gave the example. If I had put something stupid in my keywords I would just quietly fix it and not post it here in the virtual town square.

Despite a recent blog post saying they'd removed pseudonyms from basic searches my surname Warren came up in a search for Warren Buffet. I had included the keyword 'buffet' which is relevant when they are pictures of a restaurant buffet with crockery and cutlery on them.  

I don't know if you have tried Shutterstock's searches. They get very good results and the keywording is very loose by comparison to Alamy's. I don't think it is too much to ask Alamy to be as good as the competition, especially when it affects our CTR. 

 

Ray, that's really strange. Alamy have turned off the pseudonym in searches now so if you've seen this happen it must be a report from before the change (a couple of weeks ago) it won't happen now *unless* you accidentally have 'Warren" as another keyword somewhere in a searchable field.

 

I just tried to see which image it was by doing an Advance search on your name (as Contributor) and "buffet" as a keyword but it produces zero hits. Have you re-keyworded it?

 

J

Hi,  yes I re-keyworded, immediately removing 'buffet' as a short term solution. I try to correct those problems as soon as I am aware of them. Although I think 'buffet' is a legitimate keyword for those two images it is probably not essential and I'd rather not damage my CTR any further. The search was between the 16th and 23rd of June so it was after the change. The blog post came out on the 12th. I definitely do not have 'Warren' as a keyword, or in any searchable field (other than pseudo) in any of my images. I have had other strange results but that was the most glaring and perplexing example. I appreciate your time and interest.  :-)

Ray

[incamerastock]

Hi,  yes I re-keyworded, immediately removing 'buffet' as a short term solution. I try to correct those problems as soon as I am aware of them. Although I think 'buffet' is a legitimate keyword for those two images it is probably not essential and I'd rather not damage my CTR any further. The search was between the 16th and 23rd of June so it was after the change. The blog post came out on the 12th. I definitely do not have 'Warren' as a keyword, or in any searchable field (other than pseudo) in any of my images. I have had other strange results but that was the most glaring and perplexing example. I appreciate your time and interest.  :-)

Ray

 

I'm just perplexed, because we're talking about a computer here. It's not random so the change would either be in effect or not.

 

The blog post came out but the actual database change didn't happen for a week or so. I remember this clearly because when it did, it broke the search in Manage Images and I started a thread on that problem. So, if you were to drill down to the actual "warren buffet" search and get the exact date I'd still put money on it being *before* Alamy actually switched off the pseudonym field (sometime around the 18th/19th I think).

 

J

[Ray]

 

Hi,  yes I re-keyworded, immediately removing 'buffet' as a short term solution. I try to correct those problems as soon as I am aware of them. Although I think 'buffet' is a legitimate keyword for those two images it is probably not essential and I'd rather not damage my CTR any further. The search was between the 16th and 23rd of June so it was after the change. The blog post came out on the 12th. I definitely do not have 'Warren' as a keyword, or in any searchable field (other than pseudo) in any of my images. I have had other strange results but that was the most glaring and perplexing example. I appreciate your time and interest.  :-)

Ray

 

I'm just perplexed, because we're talking about a computer here. It's not random so the change would either be in effect or not.

 

The blog post came out but the actual database change didn't happen for a week or so. I remember this clearly because when it did, it broke the search in Manage Images and I started a thread on that problem. So, if you were to drill down to the actual "warren buffet" search and get the exact date I'd still put money on it being *before* Alamy actually switched off the pseudonym field (sometime around the 18th/19th I think).

 

J

And you'd win the money!  If they didn't switch it off until the 18th and not the 12th like the blog post would suggest the mystery is solved. The search was a few days before that. Thanks for the valuable information.

Regardless of the miracle and/or epiphany we have witnessed here today ladies and gentlemen I still maintain this search engine does not stack up against shutterstock or istockphoto/getty which can't be helping the CTR and Alamyrank we live by. 

Thanks J

[Ed Rooney]

Yesterday My Alamy showed 1 sale and 6 zooms and a CTR of 0.59. This morning I have 3 sales and the same 6 zooms and a CTR of 0.54. Is this information useful to me? I don't have a solid opinion on this, but the numbers I'm most interested in are sales and price. 

[Shergar]

It looks like the Huff Post has dropped Alamy from all its editions now, could this be hitting the Alamy CTR. I think the loss of the HP is probably hurting us all .