If you’re a Getty contributor and your sales and revenue have been declining, it may be time to do some searches on Gettyimages.com as a customer would search. Input some of the generic keywords that a customer might use to find your images. See where your images fall in the search return order.
I recently did some searches for popular subjects and discovered that certain brands tend to be very dominant in the search return order. Obviously, it depends on which keywords a customer actually uses. I you have used very specific keywords, and the customer thinks to use those exact keywords, your images may come up near the top. But, most customers tend to start their search with generic keywords.
I started my analysis by searching for “Woman Office.” While the default search is for Royalty Free images, I changed the search parameters to search for “All.” This way I got both RM and RF. Currently, searching for “All” gives the customer about 50% RM and 50% RF in the first 500 images. The following are the number of images from each brand that appear.
| |
1st 100 |
101-200 |
201-300 |
301-400 |
401-500 |
|
Total |
| |
Images
|
Images
|
Images
|
Images
|
Images
|
|
|
| Hero |
28 |
23 |
23 |
20 |
19 |
|
113 |
| Stone |
17 |
19 |
20 |
17 |
19 |
|
92 |
| Taxi |
29 |
12 |
10 |
16 |
12 |
|
79 |
| Iconica |
4 |
17 |
18 |
17 |
12 |
|
68 |
| Caiaimage |
15 |
11 |
9 |
17 |
8 |
|
60 |
| Digital Vision |
2 |
3 |
6 |
8 |
11 |
|
30 |
| E+ |
|
10 |
9 |
|
10 |
|
29 |
| OJO |
2 |
3 |
1 |
4 |
2 |
|
12 |
| Cultura |
1 |
2 |
1 |
1 |
3 |
|
8 |
| Hoxton |
1 |
|
1 |
|
|
|
2 |
| Blend |
|
|
1 |
|
1 |
|
2 |
| Moment |
1 |
|
|
|
|
|
1 |
| Photog. Choice |
|
|
1 |
|
|
|
1 |
| Riser |
|
|
|
|
1 |
|
1 |
| Archive Photos |
|
|
|
|
1 |
|
1 |
| Taxi Japan |
|
|
|
|
1 |
|
1 |
| |
|
|
|
|
|
|
|
| |
100 |
100 |
100 |
100 |
100 |
|
500 |
What I find interesting is that Getty has images from almost 300 brands on its site. Yet, a huge number of these brands have no images shown at all in the first 500 images. If customers use more specific keywords they might get to the images of other brands.
Trying to drill down by using more specific words often doesn’t work very well. When customers use three, four or five words in a search they get all the images with any one of those words as a keyword. Thus, instead of narrowing the search, it delivers more images although the images that have all the words listed may come up first.
Default RF Search
The default RF search for “Woman Office” is even more interesting. Half of the images shown in the first 100 come from a single brand and 88% came from just three collections.
| |
1st 100 |
| |
Images |
| Hero |
50 |
| Caiaimage |
28 |
| E+ |
10 |
| OJO |
5 |
| Digital Vision |
5 |
| Moment |
1 |
| Hoxton |
1 |
| |
|
| |
100 |
Maybe, it varies with subject matter. I did searches for “Travel,” “Medicine” and “Soccer,” and used the “All” option so I should have about 50% RM and 50%RF. The pattern is very similar to the first search for “Woman Office.”
| Travel |
Images |
|
Medical |
Images; |
|
Soccer |
Images |
| All |
|
|
All |
|
|
All |
|
| Moment |
21 |
|
Stone |
32 |
|
Stone |
26 |
| Taxi |
17 |
|
Hero |
20 |
|
Caiaimage |
25 |
| Stone |
14 |
|
Caiaimage |
9 |
|
Digital Vision |
13 |
| Caiaimage |
11 |
|
Digital Vision |
9 |
|
Taxi |
8 |
| Digital Vision |
10 |
|
OJO |
7 |
|
Iconica |
7 |
| Hero |
7 |
|
Taxi |
6 |
|
Hero |
6 |
| Photonica |
7 |
|
Cultura |
4 |
|
Photonica World |
5 |
| Cultura |
3 |
|
Iconica |
4 |
|
Photog. Choice |
3 |
| OJO |
3 |
|
Moment |
3 |
|
Moment |
2 |
| Iconica |
2 |
|
Science Photo Lib . |
2 |
|
EyeEm |
2 |
| EyeEm |
1 |
|
Blend |
2 |
|
Westend61 |
1 |
| ImageBank |
1 |
|
Ikon |
1 |
|
Image Bank |
1 |
| Blend |
1 |
|
Image Bank |
1 |
|
Johner |
1 |
| Riser |
1 |
|
|
|
|
|
|
| Room |
1 |
|
|
|
|
|
|
| |
|
|
|
|
|
|
|
| |
100 |
|
|
100 |
|
|
100 |
Why Is This Important?
If you’re submitting images directly to Getty then the collection in which the images are placed becomes important. If you are photographing the most popular subjects placing those images with a smaller agency or production company that distributes those images through Getty, as well as a number of other distributors around the world, you may not be seeing much in the way of sales through Getty.
It is also interesting that a very high percentage of the images being shown by Getty come from two production companies – Hero Images and Caiaimage.
Getty acknowledges that they no longer use a rigid collection-based grid as they once did.
They say that search results are now based on a combination of Relevance to the terms a customer enters, Recency, and Customer Interaction.
The “Customer interaction” could explain some of why these two companies so dominate the search return. Caiaimage, headquartered in London, and Hero Images in Calgary are both very active producers of high quality, in-demand images. Thus, they always have images that fall into the “most recent” category and get near the top of the search-return-order. As customers purchase more and more of their images, not only because they are good, but because they are the first ones shown, then their collection as a whole has more “Customer Interaction.” They then get a higher and more dominant position in the search-return-order due to the customer interaction with their images.
If you are a photographer producing images that you hope to market through Getty Images, it may make more sense to submit them to Caiaimage or Hero Images first and see if they will accept them. The images they accept are likely to get a better position in the Getty search than if they are submitted directly to Getty. Of course both of these production companies take a percentage of whatever they get from Getty, but if your images are not seen they are unlikely to generate any revenue at all.
It is also important to remember that Getty can change their search algorithm at any time. In the past other brands used to be high in the search return order. Now, few, if any of their images are found in the first 500 shown to customers.
Reader Input
After I first published this story I received some additional input from readers that is helpful.
It has been pointed out that the brands that come up first in searches vary depending on the IP address of the customer. Thus, customers in Germany using the same search terms as customers in the U.S. will see a totally different set of images from those shown to customers in the U.S. The German customers will see more German produced images. This may be true of other countries, we’re not sure.
Thus, a photographer doing search in the U.S. may find his images showing up very will for U.S. customers, but they may not show up at all in Germany or the UK.
There is also some speculation that IP information may be used much more specifically. For example, if it is determined that a customer using a particular IP address has purchased images from a particular brand in the past, a higher percentages of images from that brand might be shown in future searches. This could even be narrowed down to images produced by a specific photographer. While this kind of thing is theoretically possible there is no indication that Getty or anyone else is using specific IP information to this degree.
Another reader pointed out that if searchers use the advance search features and multiple keywords they can get a much more targeted group of images. Getty Images has some particularly good advance search options. These only work if the images have been well keyworded.
There have also been reports that despite the availability of these advanced search features very few customers actually use them.