Everyone accepts that iStockphoto is the microstock industry leader. Using Getty Images' quarterly figures, it's possible to estimate iStock's 2007 gross revenue at around $55.55 million. (See Royalty Free Trends At Getty for how I arrive at this number.) Thomas Weisel Partners estimates iStock will generate over $95 million in 2008. We also know they had 17.55 million downloads, which enables us to determine that the average return per download was $3.17.
Photographers considering other sites need to determine how they will compare with iStock. Various successful photographers writing on microsite blogs and independent forums offer their personal numbers. These are useful, but biased, since they represent one photographer's experience.
It would be helpful to know the total number of annual downloads for a given site, and/or the average return per download or the gross revenue of the company, but this information is not disclosed. However, the other sites (except for Shutterstock) tell us how many times each particular image has been downloaded.
In an attempt to learn more about comparative sales, I decided to use the same keywords to search several sites, order the images by number of downloads and count the total number of downloads for the 100 most downloaded images.
The search term I chose was "woman and telephone." Since this is a subject in high demand, I expected to find a significant number of images on all sites relating to these keywords. The first column in the chart below is the total number of images on the site. The next column is the number of times the image with the highest number of downloads was used. The third column is total downloads for the 100th image in declining demand.
Obviously, there were many images on each site that had fewer downloads than the 100th image, but I did not attempt to count the total downloads for all images using these keywords. The final column is the sum total of downloads for the first 100 images on each site.
| Company | Total Hits | Download on | Downloads of | Total Downloads of |
| | First image | 100th Image | Images 1 thru 100 |
| iStockphoto | 12,015 | 3,923 | 358 | 71,196 |
| Stockxpert | 5,000 | 1461 | 213 | 42,610 |
| Fotolia | 5,571 | 364 | 36 | 6,058 |
| Dreamstime | 9,196 | 201 | 29 | 5,430 |
This is not a perfect comparison. The strategy doesn't take into account how long an image has been on the site. Since iStock has been operating longer than the others, it is reasonable to expect the company to have more downloads. I can partially solve this problem by doing the same search every six months. At that point, Â I will be able to measure the six-month growth, which will be a truer reflection of relative size. Thus, not too much should be read into these first numbers.
If eventually I can gather enough data to conclude that one site has four times the number of downloads in a six-month period than another, and we know the total number of downloads of at least one, we can begin to draw some general conclusions about potential downloads by the other site. This information coupled with other individual photographer experiences should be helpful.
The numbers are also thrown off by the fact that both iStock and Stockxpert "normalize" the keywords, while Fotolia and Dreamstime do not. A customer who uses either "woman and telephone," "woman and phone", "woman and cellphone" or replaces woman with women in either an iStock or Stockxpert search will always get all the same images. The search engine automatically searches for words with similar meanings to the word used by the customer. Fotolia and Dreamstime don't use this strategy. These two sites provide a different count with each combination of words.
Normalization is definitely a benefit. Many customers will move on to the next site, rather than think to use both plural and singular or try to think of every word that might have a similar meaning, and thus miss some relevant images. It would also be wise to keyword more extensively for those sites that don't normalize.
Another factor is that certain subjects may be more popular on one site than another. Given that, it might be useful for photographers with a specialty to use the technique described above with keywords appropriate to their specialty. Determine the number of images in their specialty on various sites and the total downloads of the top 100.
This still doesn't solve all the problems. It is easier to get images on some sites than others and prices for downloads and royalty rates vary. The photographer really needs to know the potential revenue that might be earned for an investment of time and resources.
Still, we believe this information will be helpful in the long-term as more photographers trying to earn a living from producing stock images explore microstock opportunities.