A Good Panda... Google confirms a softer, better targeted update

[whatson]

Finally a more targeted panda update. One that is intelligent enough to leave good sites out of the "penalty".

I was nearly going to post this yesterday, as wife's blog was suddenly ranking well in Google, but I couldn't find any posts about Panda anywhere, so wasn't sure if that was the case. But it would appear so, Google are refining their Panda algo, so other sites, that may not have the best design, etc, still dont suffer. [searchengineland.com...]

[Robert Charlton]

Thanks to Barry Schwartz for several reports on this...

Google Confirms Panda Update Is Rolling Out: This One Is More "Finely Targeted"
Jul 18, 2013 at 1:22pm ET by Barry Schwartz
[searchengineland.com...]

Google confirmed with us that a Panda update is being released and said:

In the last few days we’ve been pushing out a new Panda update
that incorporates new signals so it can be more finely targeted.

In an earlier article, also on July 18, [seroundtable.com...] Barry had cited WebmasterWorld's current SERPs thread as possible indication of an update in progress...

There is renewed chatter in the WebmasterWorld forums
about another shuffle taking place in Google. The consensus
is that this update is likely Panda related.

Discussion here is that some sites are seeing recovery....

Google Updates and SERP Changes - July 2013
http://www.webmasterworld.com/google/4589243.htm [webmasterworld.com]

Barry also noted earlier Matt Cutts comments in his May 13, 2013 video suggesting that this would be a "softer" Panda.

In our thread about the video here...

Google's Matt Cutts: What To Expect In The Coming Months
May 13, 2013
http://www.webmasterworld.com/google/4573600.htm [webmasterworld.com]

...tedster had commented...

Matt Cutts posted a rare overview of Google's plans for the next few months for orgnic search - what website optimizers and webmasters can expect. What to expect in SEO in the coming months [mattcutts.com]

Key targets:

1. Stop advertorials that pass PageRank
2. Improve SERPs that are traditionally more spammy (adult, for example)
3. Better link analysis to deny value to link spammer
4. New hacked site detection methods and better webmaster communication about them
5. Detect true authorities better in various niches - better authority signals that could help moderate Panda impact
6. Improvements planned for host clustering - to make overly dominant results less common

Matt said this is a rough snapshot of the potential that's in the works. It's not a guaranteed promise because things can always change as these projects evolve.

(My emphasis added in the above re the Panda signals.)

[Whitey]

Around 5:08 of the video [youtube.com...] Matt Cutts said that the new Panda would look for those "additional authority signals" that would soften the effects of Panda.

What authority signals could have been overlooked. Any ideas of what those might have been?

[ Although I don't know the process, it should be noted that the rollout appears to have commenced 3-4 days ago, and according to MC should continue for approx. 10 days in total].

[EditorialGuy]

What authority signals could have been overlooked. Any ideas of what those might have been?

EMDs seem to be doing pretty well. Maybe even better than they were a few days ago, even in cases where the pages have little value (and little on them).

[tigger]

A question for those that may know - If this new Panda is rolling out does this mean to say the the current UK serps haven't been effect yet? as I'm seeing no improvement as all

[stokelake]

Anyone seen changes in uk SERPS yet?

[aristotle]

the rollout appears to have commenced 3-4 days ago, and according to MC should continue for approx. 10 days in total

Does this mean that the rankings of a particular site gradually trend in one direction, either up or down, for 10 days before finally stabilizing at a new level.

Or does it mean that each site makes a sudden up or down jump to its final level, but this happens to different sites at different times over a 10 day period?

[Jane_Doe]

I guess I was getting hammered on the "well designed part" on some of my oldest sites these past couple of years.

[EditorialGuy]

I think that, in effect (not necessarily as a goal), there were two aspects to Panda:

1) Negative signals (thin content and other sins that have written about by every SEO guru).

2) Positive signals (massive numbers of inbound links, a minimum PR threshold, or whatever) that insulate megasites, including the larger UGC sites, from Panda's effects.

Under this scenario, small and medium-sized sites with negative signals would get hurt by Panda, while the sites with positive signals--i.e., the megasites--would float up higher in the rankings if only because other sites were dropping.

BUT...

For the search terms that I watch, some of the megasites were slipping back to more reasonable levels even before this latest Panda update. (In one search that I do fairly often, Wikipedia and one of the big UGC sites have been down in positions six and seven respectively for a while, with the #1 result being an About.com page and two of the other top five results being extremely thin pages on EMD sites. So far, the only effect of the latest Panda update has been to push the "official" page for the query down to position number 10.)

My gut feeling:

Google may have been trying to reduce the "megasite effect" in its SERPs (at least for informational queries) even before the current update, but it's having trouble plucking the good stuff from what's left over after Wikipedia, TripAdvisor, etc. have sunk lower in the results. Author reputation and authority ("AuthorRank") could help with this, but it could be a long time before Google has AuthorRank figured out. (Until then, Google might want to dial down the weight that it gives EMDs.)

[purplekitty]

Other than backlinks, what other signals does Google use to determine authority? Is there anything that Google has confirmed as being used?

[Awarn]

I think they were using blog posts. What other people posted in blogs linking to your site. I had a ton of links and it looked that was just devalued. However it seemed that blog posts were viewed positively. Probably a social authority signal. What gets me is I am seeing ecommerce sites with 30%+ of their products out of stock. They look like an authority but in reality it is more like they are spamming because they rank for a product but don't have it. It so bad anymore I have customers asking me if I am legit because the sites above me don't have the products but I do. Google's definition of authority is flawed but they are improving. I still believe age is the best proof of authority. If you do good the website grows and prospers. I am surprised we aren't seeing some form of positive from having a mobile version or a responsive design.

[IanCP]

Yes, well for the present I'll still regard the last two days as just another one of those "blips".

Then again maybe I'm seeing some return to sanity. One swallow doesn't make a summer though.

[EditorialGuy]

I am surprised we aren't seeing some form of positive from having a mobile version or a responsive design.

Google recently suggested that providing a lousy mobile experience could hurt your rankings in mobile search but not in desktop search. That's reasonable. (It wouldn't make sense to use the existence or non-existence of mobile versions/responsive design in desktop or even tablet searches. If I'm searching for "widget" or "widgetonia" from my desktop computer or iPad, I couldn't care less what example.com looks like on a feature phone or smartphone.)

[Awarn]

I can understand what your saying but if your using a mobile device and there is no mobile version it can be a poor user experience. Now look at Googles recent earnings per share. The big stated issue as why Google didn't meet street estimates was the fact that mobile advertising is too low. Too low because mobile doesn't convert so advertisers pay less for mobile ads. So a good mobile site with a streamlined checkout is in fact a positive.

[jojy]

Will this update revert back penalty from innocent sites?

[whatson]

From what I can determine is that perhaps there are some new signals they are analyzing to check if a site should be in Panda.
For example, if a site was before in Panda, I think it still is, but if the site hits these extra signals then it's Panda "penalty" is reduced.
Does that make sense?

[EditorialGuy]

I can understand what your saying but if your using a mobile device and there is no mobile version it can be a poor user experience

Yes, and that's why Google is taking the mobile experience into account for mobile search.

From what I can determine is that perhaps there are some new signals they are analyzing to check if a site should be in Panda.

Yep. That's exactly what Google said in announcing the update:

"In the last few days we’ve been pushing out a new Panda update that incorporates new signals so it can be more finely targeted."

One question that hasn't been answered is whether Panda, like Penguin 2.0, will become more "granular" as it continues to evolve. From a user's perspective, does it make sense to lower rankings across a site because Google thinks one part of the site is "thin" or otherwise not up to Google's standards? Mightn't it be more reasonable (and better for users) to make quality judgments at the page or directory level?

[Whitey]

@whatson - It makes sense that the thresholds have been moved more favourably. But what those signals could have been has got me perplexed. One thing we do "think we know" is that UI was a factor, and that judging by the very poor standard of some big brands UI, "brand" ran over the top of that threshold.

Brands = authority.

What exactly has Google now taken note of that elevates "borderline" sites into that category I have no idea. Some factors to consider might be:

- the number of times a sites brand name is typed directly into the search
- the number of times searchers respond to an autosuggest result containing the sites brand name
- the relative popularity of those terms above
- the generalised quality of the UI [ e.g. factors 1 -10 ]
- differentiating content features [ no. of unique category titles versus common duplicated titles ]

Really I have absolutely no clue and was hoping for folks to scratch their heads a little and contribute their thoughts. At the end of the day it's pretty academic, and the game-plan for siteowners / SEO's and marketers hasn't changed. Focus on being useful.

On the numbers side, it would be good if folks would be prepared to share some metrics. Sites I look at over a range of languages, industry verticals, and location have increased week on week , by between 50% and at the higher end 350%. The norm appears to be around 150%. NB This roll out probably has 5 or so days to go, and I have no idea of how that will play out.

Given the drop in traffic over the past 2 years or so, I'm wondering if these types of increases provide much respite and incentive for folks to reinvest in their content. SEO is a very unreliable marketing channel - very, very.

And with technology moving so fast and diversely [ e.g. devices , applications - all requiring added resource to support ], Google would have to be wondering if there is enough good quality content to encourage a large enough site owner base to upgrade their content diversity. In my view, Google has to come up with strategies to encourage site-owners to invest in their SEO business, or they simply will have a polarised few. Simply, Google may have been too aggressive and the selection criteria previously used for quality assessment was equally aggressive.

Without diversity, in a fast moving technology driven environment, Google's product of search will lessen in relevancy to alternatives, unless they can come up with significantly improved introduction of good quality, diverse results that are relevant. They need to apply the incentives to respond well amongst the smaller end of town IMO.

[virtualreality]

I see some traffic recovery with the "Good Panda" but I dont understand where it is coming from. I dont see keywords movement or improvement. Just boost in traffic starting July 18. Anyone else experience similar thing?

[chalkywhite]

The issue I have is I was hit by panda and it took a year but I recovered, once I had done I was scared to change the site to much incase I was hit again...alas may the 9th it happened.

The way google is currently set makes webmasters loathed to "improve and update" content.

[ubound]

I was hit on May 9 too, the Good Panda returned my rankings to where they were before June 2012 (before Panda hit that is) but the traffic recovered only to pre-May 9. Weird!

[aristotle]

I didn't think any of my sites had been affected by Panda, but now see that one of them had about a 30% boost in traffic on the 18th and continuing on the 19th. This site had been slowly losing traffic for a year or so, but I thought it was because of the increasing dominance of big sites, big organizations, and big brands. But now it looks like Panda may have been involved as well.

Anyway, it's good to see that Google has finally recognized that there are problems with its search algorithm and is starting to make efforts to fix them.

[virtualreality]

@aristotle, do you actually see improvement in your keywords placement or just boost in traffic?

[aristotle]

@virtualreality
Unfortunately I haven't been monitoring the rankings for this site recently, but since the Google traffic increased, some rankings must have improved, unless all of the increase is from long-tail.

[EditorialGuy]

And with technology moving so fast and diversely [ e.g. devices , applications - all requiring added resource to support ], Google would have to be wondering if there is enough good quality content to encourage a large enough site owner base to upgrade their content diversity. In my view, Google has to come up with strategies to encourage site-owners to invest in their SEO business, or they simply will have a polarised few. Simply, Google may have been too aggressive and the selection criteria previously used for quality assessment was equally aggressive.

A trio of thoughts:

1) Wouldn't it be more productive, from Google's point of view, if site owners invested in content instead of spending their money on SEO?

2) For commercial topics, the problem for independent vendors (IMHO) is that Google Web Search is about indexing and ranking content, not indexing and ranking businesses or deals. A site like Amazon.com or Booking.com, which is packed with user reviews, will have an inherent advantage over a mom-and-pop e-commerce site that lacks a megasite's ability to attract user-generated content. By the standards of Google Web Search (and from a Google user's point of view), a typical content-rich page from Amazon should rank higher than the equivalent page from a small e-commerce site that probably contains little more than boilerplate text. From Google's content-oriented perspective, the boost given to sites like Amazon.com or WalMart.com or Booking.com isn't a bug, it's a feature.

3) For informational topics (my stomping ground), Google's standards for quality assessment haven't been too aggressive. If anything, they haven't been aggressive enough. The challenge for Google is in how to judge quality algorithmically. So far, the results haven't been terribly successful, but initiatives such as authorship markup (which provides an element of accountability) and AuthorRank (which has the potential to allow more targeted quality assessment) could improve Google's ability to pluck nuggets from the cesspool--not this week or next month, but possibly in the more distant future.

[diberry]

but since the Google traffic increased, some rankings must have improved, unless all of the increase is from long-tail.

I'm not so sure. I'm carefully monitoring a site that's slowly lost Google traffic over the past year. It's partly user behavior (that Google isn't where my audience is finding my topic, and also that mobile's getting bigger and my audience uses it... very differently for search, I'm still trying to parse this). I can't find any long tail changes that would account for my ups and downs. I think it's just that being #1 in Google ain't what it used to be on some topics.

[whatson]

@Whitey - I agree but what about domains/brands that are generic search terms as well? Perhaps the variance was so low it didn't matter.

I think UX/UI is one of the major Panda factors. Chrome has enough market share, that they can get rely on the data it creates, and we can only begin to speculate one what it might be reporting. The fact is that we know they are, and so why wouldn't they use that data to improve their index? Of course they do.

So I am not even sure if there are any on page factors for Panda, although indirectly they may result in lower Panda scores, e.g. lots of ads, or poorly written weak content may result in less time spent on site, more clickbacks, etc, and thus effect Panda score.

[turbocharged]

Wouldn't it be more productive, from Google's point of view, if site owners invested in content instead of spending their money on SEO?

It would be more profitable for Google to have more Adwords users.

A site like Amazon.com or Booking.com, which is packed with user reviews, will have an inherent advantage over a mom-and-pop e-commerce site that lacks a megasite's ability to attract user-generated content. By the standards of Google Web Search (and from a Google user's point of view), a typical content-rich page from Amazon should rank higher than the equivalent page from a small e-commerce site that probably contains little more than boilerplate text. From Google's content-oriented perspective, the boost given to sites like Amazon.com or WalMart.com or Booking.com isn't a bug, it's a feature.

Many of Amazon's reviews are not accurate and/or are for the completely wrong model of product being sold. Buyers of products that Amazon ships have an extraordinarily difficult time getting support, whereas consumers purchasing their goods at a mom and pop can normally just pick up the phone. Also, let us not forget how Amazon hijacks products from those who no longer want to sell their goods on Amazon. For years that product will be active, despite the fact that the manufacturer quit selling it on Amazon and Amazon can no longer supply it to their visitors.

For informational topics (my stomping ground), Google's standards for quality assessment haven't been too aggressive.

I would agree. Doing a basic search for exterior drain pipes in Google, I saw a listing ranking in the top 3 for Ask (the search engine). Clicking on it I went to the Ask's answers site, where the answer quoted is linked to a Yahoo answers page. Ironically, all of this clicking around to find the source and it was wrong. The holes in an exterior PVC stormwater drain pipe are pointed down and not up as the marked "best answer" in Yahoo noted. Regardless, I normally never rely on one source because I have found that the internet is the greatest source of misinformation a person can use - if not researched properly.

[backdraft7]

Who needs targeting when you're dropping a daisy cutter? Completely wiped out over the past 4 days.

[Whitey]

@backdraft7 - sorry to hear that. Being objective, if you can under the circumstances, what do you think might have triggered the fall, when others are largely responding with positive news?