Panda 3.9 Rolling out 7/24/2012

[netmeg]

Tweeted by Google:

New data refresh of Panda starts rolling out tonight. ~1% of search results change enough to notice. More context: [goo.gl...]
[searchengineland.com...]

[Hope_Fowl]

How many 1% is that now?

[Hope_Fowl]

Hmm. I click Submit once, and get two messages posted, with no edit option.

[indyank]

askthebuilder.com and daniweb.com

I am not sure about askthebuilder.com as I am not that familiar with that niche but I have not found any good answers for tech queries, whenever I land on daniweb. It is only stackoverflow where I get all the good answers...

I will have to admit that I rarely land on daniweb for my tech queries but so far my experience hasn't been good at all. So perceptions on quality differ.But ultimately what matters is the "Google quality".

[Shaddows]

Anyone know if thats
"1% unique queries"
"1% of all queries performed"

If its the former, that will be SUBSTANTIALLY more than the latter.

In any case, given the intended target, its pretty obvious that active SEOs will be seeing the most impact.

Edit - @Hope_Fowl you have only posted once, but you are onn the default 30-posts per page. Your post is the 61st, so appears at the bottom of page 2 and the top of page 3. It's normal, if not consistant

[netmeg]

pages with a high percentage of similar content cause a considerable issue to Panda.

Not in my experience.

[chalkywhite]

I honestly cannot tell whether Ive been hit again or whether it just "google these days", one day is never the same.

I hope the two refresh thing holds some water as I made ALOT of changes this month. The content is good ( I document walkthroughs of IT migrations and issues that I work on - microsoft reference them etc ), I dont know where to go or what to do if this does not bounce back soon.

[Lenny2]

Zeus, let me know if you plan on giving up... I might be interested in buying your site. Sounds like an interesting project... Lots of upside.

Chalky, hang in there! Itll probably bounce back in a rendition or two. I wouldn't change anything unless it helps the site overall, regardless of panda.

This has been a quiet update... We are basically even... Which Is a good thing, given that most pandas come with a slide to our site.

We fixed some systematic grammar and spelling errors during June... Saw an improvement in the normal SERPS no change with this latest panda update.

[Lenny2]

Zeus, let me know if you plan on giving up... I might be interested in buying your site. Sounds like an interesting project... Lots of upside.

Chalky, hang in there! Itll probably bounce back in a rendition or two. I wouldn't change anything unless it helps the site overall, regardless of panda.

This has been a quiet update... We are basically even... Which Is a good thing, given that most pandas come with a slide to our site.

We fixed some systematic grammar and spelling errors during June... Saw an improvement in the normal SERPS no change with this latest panda update. Onward we truck.

Here is an interesting tidbit for all panda sufferers:

I created a unique, hi quality landing page for a key term set that has 15m websites competing for the term. A generic key term, with not a ton of traffic, but enough... Anyway, it ranks on the first page, top 5 places on my panda hit site. The point being: apparently it is possible to rank well with new pages on yr panda hit site... Took about 1 month to rank, with only 1 internal link pointing to it. Other point, maybe google is getting better at discerning and ranking higher quality pages without depending so much on link equity.

[petehall]

@netmeg if you have not experienced this, perhaps you will at some point. Although if you have a method which works for you that allows repeated content then I congratulate you, as it makes page creation so much easier.

[jinxed]

If things stay like this, my Panda'd site will have seen a 350%+ increase since the first hit.

[fashezee]

I cant find that tweet; what tweet account was it done by?

[wingslevel]

Hey jinxed - congrats!

curious how much of that 350% would you say came from 3.9?

[robzilla]

I cant find that tweet; what tweet account was it done by?

Official @google account, see: [twitter.com...] (and usually retweeted by @MattCutts).

[jinxed]

@wingslevel

I'm seeing a 15% increase in comparison to last week’s numbers so far.

[lizardx]

Shaddows: Improving the ranking of good sites is their bread-and-butter. That's the main algo.

The negative scoring* parts target the people who "fake" quality.

the same light has gone on for me, though this was what seemed intuitively obvious. Some recent tests I ran which nobody here or elsewhere has mentioned and which I am going to keep to myself so the google guys can keep laughing, as they surely are while reading these threads, anyway, those tests totally have confirmed what you note to me. In my opinion, Panda is far more simple than I thought, but that fits the general notion of occam's razor when it comes to massive page analysis and ranking, and subsequent low quality flags being added. Google has been saying this all along, re quality of content and low quality content but I suspect their words misdirected people to much more complicated scenarios than the simple stuff google is actually probably doing, and thus the amused snickers behind google's doors.

[themaninthejar]

I love the mental image of an amused Snickers...

[zeus]

Ok I dont see anymore at all, so I think this update did not bring my site back, well I can not change anything anymore, so I have to wait and maybe sell the car now.

[Web_speed]

seeing +/- 120% increase in traffic over three sites i monitor. Traffic quality down the tubes though. No (or hardly) any sales. Window shoppers at best, high bounce rate plus lots of "wrong" country hits.

[garyr_h]

@petehall, I think Panda is much deeper than similar content. I fixed that problem around January and still have no recovery (actually, it has gotten worse and worse). I'm starting to believe it has more to do with site structure, user engagement, and things of that nature. Perhaps similar content has something to do with the user experience aspect, but I don't believe for one second that it is the main cause of many people's problems.

[indyank]

Anyone know if thats
"1% unique queries"
"1% of all queries performed"

If its the former, that will be SUBSTANTIALLY more than the latter.

I think it is the latter always...

[petehall]

@gary_h you are correct site structure is also a huge factor. In addition, fragile rewrite structure can result in some quite nasty surprises.

[menntarra 34]

I have the following experience from the last week:
Friday: -10% traffic drop till Tueasday, when i recovered 5%.
So now alltogether i have -5% traffic drop :/ . The Panda refresh definitely started on Friday, i wonder why they tweet about it 5 days later.

[Martin Ice Web]

Other point, maybe google is getting better at discerning and ranking higher quality pages without depending so much on link equity.

Lenny2, if that is true why should google send out link warnings?
Because it is a very important key for them to rank websites. It´s the only rankingfactor in regard of user action they really they really could analyse besides onpage factors. All other user metrics would never give a real picture from the website because they are not able to figure out the real traffic of the site ( direct access, other search engines, facebook, twitter.... )
Even if you take the data from chrome you couldn´t say anything about user metic without making a big percental failure.

Just my two pence:
Besides all this discussion there is only two winners in this game. 1st: google - more revenue since panda, while poeple do not find any professional pages with deep content anymore but through adsense

2nd: amazon because the site ranks for nearly ever query with multiple pages and Cutts told us that he is a fan of amazon.

[chalkywhite]

Traffic remains the same by the looks of it , however impression in WT is up. Strange.

[Shaddows]

@menntarra 34

The average rate of algo changes is more than one per day. Not every ranking drop is Panda/Penguin.

[zeus]

I have a whole new theory to Panda, which also sounds logic. Im not sure google can judge if you have good content or not, thats why Panda is maybe 85% of what a user does on your site with a google tool activated, so if a user dont surf your site or leaves in less then 2 sec. thats a negative panda point, which is collected in your domain panda box.

So thats why :

- Panda update is every Month, so enough new points is added to your domain box to see if you have more positive points now, then negative.
- Thats why you dont see any boost right away when you have made changes on another Panda update.
- Thats why a domain change or subdomain gives you a quick panda relief.
- Thats also why no one have a 100% clue what has to be changed to get out of Panda, be cause a visitor gives you the points, "not what you change on your site", ok if you make you site better for the user its just positive.

[rango]

I'm with you on that theory zeus. I spoke to someone the other day who used to work at MS and is familiar with some of the stuff going on in the Bing algorithm. Click trails played a big part in the way they were doing things (2 years back already).

I'm convinced Google does similar stuff to generate the Panda results, just more advanced. And they're more willing to ban the whole site for one sub-par part. That feels mostly like a cop-out on their part to me.

[Shaddows]

thats why Panda is maybe 85% of what a user does on your site with a google tool activated

Click trails played a big part in the way they were doing things

For UX devotees, Claaarky started a thread on Exit Rates [webmasterworld.com] which covers most of the key point, for and against.

[petehall]

- Thats also why no one have a 100% clue what has to be changed to get out of Panda

You are so wrong. Show me your site and I'll show you the way.

[muzza64]

"Show me your site and I'll show you the way."

petehall, is that an open offer? I would be very interested in any help you can offer.