In the latests Search Off the Record, Gary Illyes and Lizzi Sassman from the Google Search Central team had Danielle Marshak, a Google Search Product Manager for Videos, as a guest on the podcast.
Danielle Marshak, who created a couple new SEO video help resources recently, spoke about how Google understands the content in videos.
Gary Illyes asked how does Google understand what is in the videos and Danielle explained they extract text and understanding from videos in a few ways:
- Audio from the video file is used to understand what words are being spoken
- Characters can be extracted from the video. This includes characters like headings throughout the video to understand where sections (important moments) of the video are.
- Visual information can be extracted, such as objects, animals or motions. Danielle said this is harder for Google to handle
- Structured data is used by Google as well in some ways
She said, still make sure to use structured data, Danielle said "don't abandon structure data just yet. We love structured data, because as we were talking about, even though we've come so far, we're still at the tip of the iceberg, in terms of being able to really deeply understand videos."
She also said they use text from the page to also understand the video.
Structured data is also used to discover the video as well, and get the video content.
One additional thing is that Gary Illyes said they may use short form video content for the Google help documents, but not sure yet, and if they do Danielle would help.
The good parts of this video starts at about 7 minutes in (yes, skip the cats - sorry Gary). You can listen to it below:
Forum discussion at Twitter.
