| 2:23 pm on Aug 23, 2012 (gmt 0)|
I would guess that revealing the patent would make it moot. Now people know that after their changes, they can expect a timeout. It all seems like a bit of paranoia to me and just more code to muck up the works, but like netmeg says, they may not (or may never) use it. Back to Panda...
| 2:44 pm on Aug 23, 2012 (gmt 0)|
In reading various threads about Panda, I've seen reports from people who have made multiple changes to their website spread out over a period of time of a year or more. To the algorithm, it could appear that these multiple changes were individual reactions to rankings movements that occurred in the interim. Also, even before this patent appeared, many of us have observed over the years that a major change to a website often causes an initial drop in rankings and traffic.
| 3:25 pm on Aug 23, 2012 (gmt 0)|
Panda: traffic and conversions within normal variance, maybe traffic up a tiny bit.
Re: Major changes. There is no way your document score will be the same after a major change. Even without the patent, it makes sense for there to be a drop while the algo measures a score they have a high degree of confidence in.
That said, since at least Panda and especially Penguin, every "fix" leads to ranking drops in the short term- a strong suggestion that the patent has been somewhat incorporated.
| 5:57 pm on Aug 23, 2012 (gmt 0)|
That's why Google's confirmation of the dates for Refreshes is useful. Any other losses or gains after attempted Panda or Penguin fixes can be disregarded until the the next announced Refresh.
| 7:01 pm on Aug 23, 2012 (gmt 0)|
@Shaddows and @aristotle: I realize that the transitional ranking period patent has its own thread at WebmasterWorld but perhaps it would be useful in discussing the Panda refresh, too: the patent mentions "random" rank movements, not only "down" which would mean that you could also see inexplicable "ups" in addition to "downs" that everyone talks about.
So, what happens after a refresh then: "downs" get reported because people are pissed and "ups" don't? Anyhow, from reading the [concise version of] the patent I got a feeling that the transitional rank is applied following a crawl of a page to each page individually and the length of the period depends on the nature of the change. Some of the changes, such as link-related ones, can lead to particularly drawn-out transitional periods of up to 70 days, if I got that right. In fact, even the shortest transitional rank period seems to be longer than the interim between some of Panda refreshes: 20 days, and that's for some really benign on-page changes.
In other words, it seems that you need to wait for a Panda refresh three months out, not the next one right after you made any changes to the site, to know with any kind of certainty that you were making the right changes.
| 11:21 am on Aug 27, 2012 (gmt 0)|
I have been able to be regain the lost rankings for my client post this panda refresh by making some changes to the SEO strategy. Thanks GOOGLE!
| 1:38 pm on Aug 27, 2012 (gmt 0)|
Can you mention what you did?
| 11:23 am on Aug 28, 2012 (gmt 0)|
Interesting: GWT now shows me a massive spike in downloaded data but not pages crawled for this release. Looks like googlebot only grabs certain portions of a page most of the time, probably omitting stuff it sees sitewide.
That would mean googlebot blocks some things from downloading from my site which would explain why it is that when I change them slightly it has no impact on serps.
| This 38 message thread spans 2 pages: < < 38 ( 1  ) |