Google Panda Purge and Merge - Do 301 Redirects Work?

[Sgt_Kickaxe]

A highly discussed "fix" for a perceived Panda slap has been the purge and merge approach, where a webmaster selectively purges low value content and/or merges low value pages together.

With regards to the merge half of that equation, is a 301 permanent redirect and content move from one low value page to a better location not essentially saying "Yo, Panda, this crappy content of doom is being moved onto that page or pure awesomeness"? That in turn might be interpreted as "hmm, the crap is on the move, step on it again" resulting in a lower value to the receiving page?

- fewer low value pages = good
- fewer low value pages via 301's from them to better pages = ?

All kidding aside, should I worry about redirecting low value content onto pages that were not affected by Panda? e.g. should purge and merge become purge and purge some more? I don't have enough data to have a definitive answer on my sites just yet.

[synthese]

What if you're not purging the right pages? You could be removing the good stuff.

Trust me - the thought has crossed my mind. I used a bunch of criteria - low page views, high bounce, and just gut feeling. Whatever didn't make the cut was gone. Where there were some page views, did a 301 to something similar (although this was not always possible).

Began this in Nov 2011, and despite a small improvement in search referrals in May 2012, on June 8 it dropped back again... WE continue to add new content, but far less than of old.

[freejung]

In my case 301 redirects have succeeded in removing most of the redirected pages from the index, but so far there has been no noticeable recovery from Panda.

[bloupbloup]

Hi,
I would like to say that i hate both 404 and 410. I tried them both and i did not see the difference. Nevertheless, i only saw that 404 and 410 harmed by Internet Traffic.
I dont know lots of things but i am sure about it.
After deleting pages, i looked in my logs. I have seen that googlebot was slow to get rid of the 404 errors. For instance, if you remove 100 pages. You are going to get 100 404 errors in your logs. After each log, i have witnessed the same behaviour from googlebot, It disappears for sometimes. It is like 404 or 410 errors slowed down crawling. Interestingly enough, The Traffic also goes down and then recover zhen there is no 404 errors anymore. In webmaster tools, 404 appears as "crawling errors". It can take up to 6 months or even one year to flush 404 errors if you have few backlinks.
Google says that 404 is OK but i dont see that in the logs

[bloupbloup]

I said before it took a while. Let s talk about my own experience. I deleted 400 tags from a website. The tags URLs appeared as crawling errors in webmaster tools. It flushed 10 tags every day. In other words, it should have taken 40 days to flush all the 404 errors. I stopped the issue by doing a 301 redirect to homepage. By the 404 is like 410. I takes a long time to get rid of those 404 or 410.

[lucy24]

In webmaster tools, 404 appears as "crawling errors". It can take up to 6 months or even one year to flush 404 errors if you have few backlinks.
Google says that 404 is OK but i dont see that in the logs

Welcome to the forums :)

It can take much longer than one year. I believe the current record is 8 years. Google never forgets an URL. But just because they call something an error doesn't mean you did something wrong. They just don't have a very big vocabulary. Anything other than a 200 is listed under Errors.

[indyank]

There are definitely several aspects to quality. These crawl errors were definitely part of technical quality along with broken links, site speed etc. There are several evidences to believe that Google does give this a certain weight in its quality evaluation as they did unearth and report several of these crawl errors for Panda victims (as early as the first release) and surprisingly several of them were for pages that were removed from the site, years ago. This, not only confirms the "Google never forgets pages and their urls" theory but also the fact that they did run a complete quality evaluation including technical quality, using whatever (unforgotten) data they had with them.

[tedster]

they did unearth and report several of these crawl errors for Panda victims

Do you mean that the "errors" show up in Webamster Tools? And if so, how do you connect that to a Panda issue?

[indyank]

Yes they did show up in bulk in Webamster Tools immediately after Panda 1.0 was released. This was for whatever sites that I had seen to have been affected by Panda. The same happened to sites that got affected by the international roll out of Panda.

Several of those reported crawl errors were for pages that had been removed years ago or for pages that never existed AFAIK.

[tedster]

I think the timing is coincidental. "Crawl errors" is simple a technical description - it doesn't mean "you messed up and need to fix this." Instead, it just means "FYI, googlebot discovered links to this URL which doesn't currently exist. Change something or not, whichever makes sense to you."

[indyank]

You may be right and such errors might not be affecting the site directly...even if they do, they might be carrying a smaller weight.

But since Panda is an evaluation of the entire site, the timing suggest they might have crawled all URLs for the site in their database (including the history) and not just the indexed ones. The crawl on all the historical URLs might have surfaced so many errors.