The Google Sandbox theory started in early 2004, where new sites would not rank well for a while. Well, Gary Illyes and his Google friends spoke about it in the latest Search Off the Record podcast. In short, Gary said he thinks it wasn’t a ranking related thing but a symptom from how Google’s index worked back then.
Gary said “if I recall correctly, we used to have this batch-based indexing system, right? And then that did have this weird effect because… how was it? I think we rebuilt the index every month, but only once every month. And then that meant that if you created a website at the wrong time, as in farthest away from when we built the index, then you had this long wait period. And in…during that time, you couldn’t do much with that site because, well, we were not indexing things from your site. And then somehow that indexing sandbox, I guess, transformed into a ranking sandbox in people’s minds as well. Which was probably not that helpful. And then we just got to enjoy talking about it for like 20 years.”
He said this at 12:16 into this podcast, you can listen to it here:
In 2017, Google said again there was not Sandbox and of course, Google did not call this a sandbox. Was it an indexing thing? I am not sure. Matt Cutts in 2005 once said there might be something like a sandbox plus there was a hack to see the search results without the sandbox filter.
So does Gary’s explanation make sense? Maybe? I am not sure. Gary did not join Google until 2011 – seven years after the Sandbox theory, so he was not working at Google during this theory.
He also added:
just remember that I wasn’t around back then so my knowledge about what really sandbox was is based on REALLY old documents i found on our intranet. https://t.co/4x7wPXBSnr
— Gary 鯨理／경리 Illyes (@methode) August 25, 2020
Forum discussion at Twitter.