Comp Use Images And Search Algorithms

For agencies that offer subscriptions, Image-on-Demand and Enterprise options, how do “Comp Use” images affect search algorithms? There is an increasing trend driven by Adobe Stock and Shutterstock Enterprise to allow customers to download a lot of images for comp purposes and only pay for the ones they actually use in a final deliverable project.

With traditional subscriptions every image downloaded is counted equally, regardless of whether it is eventually used. Even, when the customer pays separately for every download a lot of the images downloaded never make it into a deliverable product. The stock agency has no way of knowing which images get used and which ones don’t, so traditionally they have counted everything equally.

Now, many of the agencies have more data and this will undoubtedly change their search algorithms. Exactly how, only the agencies know and they aren't telling because this is proprietary information.



In the Enterprise environment images also sell for a variety of different prices depending on the user and the use. One image may sell for $10, another for $100 and still another for $500. Does the total revenue generated by an image affect its position in the search return order, or is it just the number of times it has sold?

Consider this. Does an image that has sold 5 times for $10 each time get a higher position than one that has sold for $100 or $500? What if the $10 has only sold twice and the others once each?

Why Is This Important?

All stock photo collections are getting huge. If images aren’t near the top of the search return order they don’t stand much of a chance of ever being seen. New images are often given some time near the top of the search return order, but given the number of images being added every day, if a new image doesn’t sell quickly it doesn’t stay high enough in the search return order to have much of a chance of being seen.

Image producers need to recognize how the changes in the way customers pay for the images they use may decrease the chances that the producer’s images will be seen by potential users. For example, at Shutterstock if you look for “happy woman in office using computer” you get 15,364 images to choose from. What’s the chance your image will be in the first few hundred.



Most stock photographers choose to photograph subjects that are in high demand because, at today’s low prices, the only way they can possibly cover their costs is for the image to be licensed many times. But, if there is not something very unique about the image, that can be described in keywords customers might use to search for the image, and the image hasn’t been used within six months after it is uploaded, there is very little chance that image will ever be seen.  

If return-on-investment is important, think very carefully about what you shoot. Be aware of the available supply of similar images that could fulfill the same need and try to find keywords customers will use which your competitors may not have considered.