This case involves a man who attended a large music festival. While the man was at the music festival, he became drunk and briefly exposed himself as a videographer was capturing a panoramic shot of the event. The camera belonged to a band that was documenting their tour. Following the music festival, the band published the video to their website and it became viral. The shot of the man exposing himself was included in the video without his permission and did not blur out his face. A former colleague of the man posted a comment on the video including the man’s name. Because of this comment, the video in question would appear on the first page of results whenever a Google search of the man’s name was performed. The man subsequently lost his job because his employer became aware of his inappropriate conduct. An expert in search engine optimization and data analytics was sought to approximate the number of views the video had based on the number of times the man’s name has been searched since the video was posted and calculate the damages as a result of those views.
Question(s) For Expert Witness
- 1. Please describe your background in search engine optimization and data forensics.
- 2. How would you prepare an estimate of the number of views these images would have received?
Expert Witness Response E-132265
This task may be a bit tricky for a couple of reasons. First, the image might have been ranking for keywords other than just her name. In addition, historical search data is not easily accessible unless someone’s been recording it. However, there may be ways to produce an estimated range. If the image has been taken down but we know the URL of the image, that can help determine a more accurate estimate. There are several sources of information to gather in this case. There are pageviews (times the image was viewed on the website) vs. times the image appeared in search results of different search engines (impressions). The kind of results (images vs. no images) Google and other search engines choose to display in search results is a moving target, as they are constantly testing and evolving their algorithm. There are also inaccuracies in data. Although tools can give estimated volume, though not historically specific, if the volume is low, they will come back with “No Data Available.” Those are some of the complexities, but it’s a deep rabbit hole and much depends on what you have access to. The defendant’s Google Analytics (or Search Console) data would be helpful, or other relevant information about the image filestructure or page source code it was on.
Expert Witness Response E-037602
I’ve known the daily pageviews of hundreds of websites since 1995, and often “time on page” — but I had unlimited access to the domains’ traffic logs. I could prepare an estimate of the number of views these images received based on 3rd-party web-tracking services. In this case, the plaintiff would have to subpoena the defendants’ weblogs from their hosting service to get the hard numbers on how many people clicked through and viewed the given pages. The search engines would also have hard numbers, but only on how many people clicked through, not on how many people actually successfully accessed and viewed the pages. The more aggressive law firms often base claims on “page views” (difficult to nail) or “potential viewers,” which of course if it’s on search engines, means “the connected world.”