Here I am, back on my BS again…
So today we are open sourcing the first thing that Bryan and I made together when he came to Local SEO Guide two years ago (he did all the hard work.)
It’s an image change detection tool, using Python, to scrape Google searches (particularly branded, local, searches) and report on if there are any changes to the image in the Knowledge Graph.
As many of you know this image is a huge drive of clicks and various forms of traffic, so it’s important to get it right and know when it changes. I can’t tell you how helpful this little bugger has been over the past couple of years. Just as an FYI, this is what lead to it’s initial creation, before Bryan beefed it up with Python.
If anybody at Google has a problem with SERP scraping well, take it up with the US judiciary and release tools that solve your advertisers (and others) problems with your platform.
Here It Is/Come and Get It
So just up top, this script does a search for a term, looks at the file in the primary position in the knowledge panel and saves it, then checks to see if it changed.
Google was ever so gracious to change the image from a file URL to Base 64 encoding at the end of February, so it’s not as easy to just look at the image and see what the change is.
This is what it looks like when running so I can have a cool terminal screenshot as my Featured Image for this post.
Check out an example output of the data at this Google Sheet here. I won’t go into column definitions or anything, as that is what our GitHub is for…
3 Response Comments
Woah wait. How are you determining which image shows for the knowledge panel? And is it similar for GMB listings?
You would think Google would automatically notify GMB managers of changes like that. It is hard enough to need to check back and approve or deny changes suggested by customers or competitors. The featured image you would not see in the dashboard, you would need to view it on Google maps.
Thanks for sharing Dan! We are hoping to use it very soon.