Many Weeks since the Panda Update - Any Improvements? [part 2]

[rustybrick]

< continued from [webmasterworld.com...] >

So, still, no one is seeing any significant improvements?

[edited by: tedster at 5:00 pm (utc) on Mar 25, 2011]

[randle]

If you are a brand name you get all the traffic,

Or put another way, if you search in Google the results you get are top heavy with big brand sites. There was a time you went to Google for information, not an index of brands. I guess thats the Panda bear's idea of "quality".

[Pjman]

I'm giving Panada another month to fix itself and then I'm going to suggest to my users to switch to Bing. The query results are really tragically poor since Panda.

Like the Pepsi challenge (taste test). I'm going to challenge my users to compare and see their results.

[My_Media]

Google sucks now, what in the world eHow has to do with health? They ranked #1 all over for lots of my medical longtail keywords.
Google! You do not send people to see a real Doctor rather send them to writer? LOL
Do they understand what is health? LOL
Do you (Google) know what is health? You should at least hire some well known doctors to go over the top searches then approves it.

Time to move to Bing.

[Whoa]

Like others, I'm a bit up relative to the post-Panda crash, but still way down relative to pre-Panda numbers. I'm now down 35% to 40% versus Pre-Panda, whereas for a while I was down more like 45% to 50%.

[ascensions]

Next Panda I see, is becoming lunch-meat.

[indyank]

and matt cutts is talking about trust factor as part of this panda update...medical advices in e-how ranking on top is total fail...

is google and matt giving a new definition to spam?

[SEOPTI]

There is a solution, get a new domain and move your content.

[crobb305]

written by ... "She holds a BA in English literature from Placename State University-Widgetville."

Does the site/articles have .edus linking in? There has to be something that is causing it to be "trusted" other than the fact that they nofollow all their outbound links (thereby inflating their own PageRank and reducing the potential for external SEO abuse).

I am finding Yahoo Answers ranking #1 for some big phrases in my industry.

[edited by: Robert_Charlton at 6:26 am (utc) on Apr 2, 2011]
[edit reason] examplified quoted text [/edit]

[Shatner]

>>There is a solution, get a new domain and move your content.

It's not a solution for me, my site is not a content farm designed solely to game search engines into giving me good placement, so my domain has value to me.

[Shatner]

Today I got all my traffic back. Returned to pre-Panda levels!

April Fools.

[crobb305]

I really am seeing some traffic throttling, up about 15% from the past 3 Fridays. Not as good as Monday when I saw a 24% improvement. I don't think there is an update occurring, probably just some flux and perhaps some re-ranking for my new content.

[Robert Charlton]

I have seen a single tweet (not even their tweet) be #1 for a very popular term.

I've seen a single tweet ranking... not a very popular term, but for a longtail phrase. I was nevertheless surprised to see it #3. The tweet was a promotional tweet by a well known art museum referring to a favorable newspaper review of one of its shows. I was searching for reviews of the show.

Google, in this case, and in other recent longtail searches I've been trying, appears to be giving heavy weight to onpage text matching, but combined with strong authority for one of the words or phrases in the query.

There is another factor in this case worth mentioning, though, and that is whether the result returned would lead me to what I was looking for. In this case, the tweet led me to two of the best online reviews of the show... one of which, I should add, had such bad onpage optimization that I wouldn't have found it any other way.

[crobb305]

hey Google here's an idea...if ehow is such a valuable resource, why don't you start ignoring their nofollow, and let their links count for those of us out here who have been referenced and duplicated there?

[Pjman]

I have seen a single tweet (not even their tweet) be #1 for a very popular term.

My tweet from 3 weeks ago ranks #2 for a term that is way out of context. I landed in Boston and said "Better hide my pinstripes, I just landed in Red Sox land!"

The term gets 1500 searches a month. G has lost their mind.

[koan]

if ehow is such a valuable resource, why don't you start ignoring their nofollow

Their links are not only nofollowed, they're deliberately generated by Javascript code, so mostly nonexistent for search engines.

[dickbaker]

Crobb305, I'm seeing throttling, too. Since the Panda ate my site, Google as a percentage of referrals is 28%, down from 44%. It's 28% +/- .75% every day. Yesterday it was 24% by mid-afternoon, but finished out the day at 28%. Today it's 25% at 5pm CST, but it will be 28% by midnight.

[walkman]

Does the site/articles have .edus linking in? There has to be something that is causing it to be "trusted" other than the fact that they nofollow all their outbound links (thereby inflating their own PageRank and reducing the potential for external SEO abuse).

For some content farms it's easy: they got so many links last month from bloggers and media saying that Google is finally fed up with them and how much they suck...so Google gave them better rankings.

[Dan01]

I really am seeing some traffic throttling, up about 15% from the past 3 Fridays. Not as good as Monday when I saw a 24% improvement. I don't think there is an update occurring, probably just some flux and perhaps some re-ranking for my new content.

I hope that is what it is.

I have seen content shoot up because of something in the news or on a late-night show. If Jay Leno mentions something, people will query it more.

[indyank]

I guess it depends on when the tweet was made. They might be ranking for the initial few days due to QDF.They might not stick there.

[graeme_p]

My traffic is up about 50% on last year, but it was up about the same before Panda.

[Roaming Gnome]

A little hope... After this last shuffle I ranked the highest I've been since March 9th for one of my bigger terms. Could be something, maybe nothing. * Shrugs *

[Mr3putt]

*sigh*

I just tried to look up the percentage decline of administrative costs that are applied to universal health care. It was a 3 word query.

I wish I was one of the people surveyed when Google asked "would you trust this site?"

I defiantly would have said no to yahoo answers.

[grimmer]

Have removed many low quality pages with 410 as tedster suggested, starting from yesterday, it feels like our site is less demoted by google, the rankings are improved across many search terms. Hope this is a sign of coming back.

[crobb305]

Have removed many low quality pages with 410 as tedster suggested, starting from yesterday, it feels like our site is less demoted by google, the rankings are improved across many search terms.

What were your criteria for deeming them low-quality? Did you use Google Webmastertools data, a particular threshold?

Also, what fraction of the low-quality pages did you leave in place?

I defiantly would have said no to yahoo answers.

Yep, I am seeing Yahoo Answers at #1 for some big phrases. The "answers" provided are just spam/scam links.

[ascensions]

I'm sure the argument is that "we" could always find examples of poor quality, but I just did a research paper and it was very difficult to find quality professional citations and reference among the Yahoo Answers type sites and stuff written by 6th graders. I've been a Google cheerleader for the last 10 years, but I'll admit... I turned to Bing to do the paper.

[grimmer]

In our case, we have a lot legacy content contain empty member profile, or content that is not related to our site, it is not a hard decision to remove those, and we just have not done it until now.

[tedster]

As we're noticing, the web has swelled with garbage, shallow content. I certainly can see why Google would want to find a way to algorithmically identify it. Doesn't mean they've got it right yet.

On the question of what content to remove, I'd start by looking at the pages Google demoted the most. You should be able to see that from your analytics package by confining the report to entry pages with US traffic from Google Search. In other words, be data driven, and recognize that there is a site-wide effect that spreads out from those pages that were most impacted. You can't buy into the site-wide effect, you need to zero in on the real culprit pages in Google's view.

[SEOPTI]

I'm not sure why some people reported new URLs on pada affected sites were not affected. Those people spread completely false information.

I tested this and did a 301 on a folder, all the URLs in the new folder have been affected in the same way.

Some people should really stop spreading nonsense. The only way to escape this mess is to get a new domain, this algo also affects ALL new URLs on a domain which has been demoted.
There is a demotion of the domain in place and who knows when the algo will re-rank again, just imagine in six months, I will not wait.

[crobb305]

On the question of what content to remove, I'd start by looking at the pages Google demoted the most.

I've been using WMT to monitor average positions. What is complicating matters is that I will see a page drop 300 positions, then start to rise; or a page will show average position of 300 (or a huge 300-position drop), but with a simultaneous increase in impressions of 20%+. Very confusing data with little to no consistency.

[SEOPTI]

Yeah, the WMT position data is a mess, I don't use it. This is no reliable source for making a decision. First they need to fix their bugs.

They are talking about AI, decision trees and other sophisticated stuff but are unable to fix minor php code. LOL, really funny!