What to do after the Penguine Update

[Ninja Commerce]

I wanted to open a thread about 'Penguin', share my thoughts and hopefully get a discussion going because firstly it's of interest to me, but secondly because I am sure such a thread will be helpful to others on here.

As I see it, Penguin was mostly about link building - and in particular unnatural link profiles, although it's hard to be totally sure since there has been a lot going on with Google lately (splog networks being killed off, accidentally classification of domains being parked etc...)

In the past Google used links for 2 reasons:

• to indicate quality of a site (authority)
• to get clues about the topic (relevancy)

And of course on page factors also counted for relevancy, but presumably not authority.

As a result site owners and SEOs have been building links using exact match anchor text, as this was the most effective way to build both authority and relevancy. As long as the links were on good sites and your on-page SEO matched with the anchor texts you were pushing you were generally good to go.

I think that the main driver behind Penguin (and other smaller changes over the past year which have been leading up to this) is the fact that if we're honest, real, natural links practically never use that exact match anchor text you might have liked them too.

Example
I mean, look at any seomoz blog post - these guys always link to other relevant resources, but how often is the anchor text an exact match?

More likely the text used is: "click here", "look at this example" or something more descriptive like "Complete Six-month Off Page SEO Gameplan".

Without realising it, by using exact match anchor texts you make it really obvious that you are trying to manipulate the SERPS, and that makes Google grouchy.

So what I think Google have done is removed (or at least vastly lowered) the use of links to indicate relevancy. Now your links are only giving your site authority, your on page stuff indicates the topic of your site and Google will probably also infer relevancy from the topic of the sites linking to you (something much harder to fake).

And at the same time it seems they have added in a filter to spot anyone who is still over doing it with exact match anchor text.

Conclusion: Action points

So my plan post-Penguin is to all but drop exact match anchor texts, in favor of links that look how you would expect them to look, using either noise terms (like "click here") or more descriptive terms that tell the reader what to actually expect when they click on the link.

Then of course it is up to my on-page SEO to communicate to Google just what each page is about and by only getting links from relevant sites that message should be re-enforced.

Agree? Disagree?

Anyway, that's my take on Penguin so far. What does everyone else think though? Anyone finding anything different to this?

It's early days of course and there's probably more to it. I have left out social media for instance, that is because I don't think Penguin really affected this, but I'm hoping there will be some other view points out there...

 

[terryuk]

At the same time I wouldn't go too much hassle to trying to beat it.

You should try underlining or bold too

 

[vickyallison]

Anchor text seems to be out, raw link volume seems to be in.
It's a sad day for small businesses that will never outrank the big retailers because of this.

 

[Ninja Commerce]

Hi Vicky.

I don't think this makes it impossible for small businesses to compete - quality of links still matters a lot, and making those links as natural as possible is the main goal.

For a small retailer I think the emphasis (as well as a good link building strategy) needs to be on differentiation. There are plenty of creative ways to do this, depending on what your niche is.

This is no different than how business has always worked though, even before the internet.

If you can design a shop that makes it as easy and as engaging as possible for your customers to shop you can out-compete the behemoth sites who sell everything.

It also amazes me though how many of these big sites still rely on directory links and article spinning etc to build links - the door is wide open for the small retailer who really 'gets it'.

 

[david64]

I think you are right with most of that, though I question how much anchored link text has the potential to unfurl spam flags. One term I have always kept an eye on is [web design]. Companies slide up and down that SERP and burn in a spate of questions from their clients about where their rankings went. All that happens is some web design companies with abhorrent link profiles get bunked off and others with equally crappy, but thinner link profiles move up.

I also think that a lot of penalties are applied manually to domains. Though this Panda gubbinz appears to have involved blanket penalties for disgusting link profiles.

I can see these updates causing a lot SEOs to shut shop and a lot of clients to become angry at their borked rankings that they thought would be a relatively long-term investment. I've already seen a few go under. This is because they are like hurricanes across the free web, depositing globs of festering link juice in their wake. If you keep your deposits undetectable, you will bob in the wake of Daddy G's meddlings.

I'd suggest any companies who have several hundred to a few thousand a month to spend on SEO to look at affiliates instead. That budget is not much for SEO, and particularly not real SEO that is not going to show any considerable return in the short-term. The majority of SEOs (though they won't admit it) are link brokers and spammers - that's all your going to get by hiring them. If you blag AffiliateWindow, they will let you in for a few hundred and then you can offload the SEO for free to careers spammers and authority sites, so long as you offer good commissions.

With this latest update it really has become about authority. Though I question how much this has to do with algorithmic judgements and suggest that Google has become the biggest human-edited directory.

The crux of the issue has been that clients want to see a return for their investment quickly. They don't want to be investing over 3 years for a white hat fantasy that they don't understand or can see returns for. SEOs want to make the most amount of money for the least amount of work. Clients don't want to hear, nor will they that there is a good chance they will get borked off the SERPs over night. Since it's been easiest to get quick results without much work with various link shenanigans, gutter SEO has prevailed. I see the next move going towards less detectable link shenanigans. This is more expensive, which is further going to shut SMEs out of the spamisphere. Genuine WH SEO wont rear it's head until vast swathes of the industry are forced under and SEO moves in-house.

 

[Andy Miller]

Agree with TT build a good user friendly web site , all my sites improved after penguin I think because other sites were demoted.

 

[nitro23456]

I wouldn't rule out exact match anchor text entirely.....

The cut-off point for penguin with regards to anchor text seems to have been roughly 60%. So sites with >60% exact match anchor text seem to have been bitten whereas sites with <60% don't seem to be as bad...... just to muddy the water there seems to have been exceptions to these % as well.

To be safe therefore I would advise aiming for approx 50% anchor match and 50% everything else all built from natural looking sources mainly on-topic but not all.

There have been sites with >70% exact match anchors that have got away with it and what possibly seems to be consistent with them is that they are relatively young sites where the first few terms on the page title matches the anchor text of the links........ suggests that there is some weight given to links that have been present for a while i.e. google are looking for people who have tried to capitalise on anchors from the last few years when the going was good.

ETA: thought Id add the source before Steve Gibson starts wittering on. http://www.micrositemasters.com/blo...need-to-act-smarter-and-5-easy-ways-to-do-so/

 

[inbound interactive]

This is really interesting.

I have a new client who have had previous "SEO work done" basically some tin pot company that builds forum signature links all day long and charges the earth for it!

Anyways I got the business as they were under performing due to the crap work being done and then all this talk of over optimisation starts.

I of course was thinking fantastic the site is going to go down hill as there dev team is really slow at implementing my suggestions to improve the on page which by the way is full of duplicate content, stuffed titles etc and as I mentioned before has a terrible link profile.

So the 27th rolls up and "penguin" strikes and nothing changes!

By this point nothing on page has been done except for major basics, homepage canonical fix, sitemap and robots.txt files added and I'm stuck waiting for dev work to begin.

Anyways to try and combat the over optimised link profile I decided to build partial match links.

I went about my normal link building, reaching out to other sites in the space and built 20 links from 10 sites, one link going to my target page and one to the about or contact page.

Traffic is up 60% month on month and 113% year on year! In my experience the site should have been nuked! But it has only improved, obviously this could simply because I AM ACTUALLY LINK BUILDING unlike the other muppets that give SEO a bad name! Or because I have gone for the partial match anchor text, I'm not sure but whatever the case it hasn't been negatively effected by penguin.

I think I will set up some stand alone tests on some more established sites and see what happens.

As for exact match? I don't think it's the end but further testing should shed some light.

 

[webgeek]

Anchor diversity that I personally find solid:

Brand (name, domain name or url chunks) - 40%
Fluff (click here, get this, find out more) - 40%
Exact match (with variations) - 20%

You can go higher EM, but just how much higher is under review [several sites are using 40% EM and doing fine at the moment, but this could change].

Oh, and when I say exact match, I do mean a ton of exact match phrases and not just one or two keyword targets.

 

[Annoying Donkey]

So my plan post-Penguin is to all but drop exact match anchor texts, in favor of links that look how you would expect them to look, using either noise terms (like "click here") or more descriptive terms that tell the reader what to actually expect when they click on the link.

Kind of agree. You'll need some anchor text to rank. Just increase the proportion of non anchor text until you see +ve results.

Then of course it is up to my on-page SEO to communicate to Google just what each page is about and by only getting links from relevant sites that message should be re-enforced.

I disagree. On page is (relatively) irrelevant to Penguin IMHO.

Anchor text seems to be out, raw link volume seems to be in.
It's a sad day for small businesses that will never outrank the big retailers because of this.

Not so. I'm seeing big opportunities where previously there were none. Just keep the anchor text links less than other, non optimised links. Focus on quality rather than quantity.

With this latest update it really has become about authority. Though I question how much this has to do with algorithmic judgements and suggest that Google has become the biggest human-edited directory.

Penguin is (mostly) about anchor text, not authority. We need to distinguish between Panda and Penguin. Panda has a human element, Penguin is an algorithm.

So the 27th rolls up and "penguin" strikes and nothing changes!

We have a test site that we believe should have been hit but wasn't. There are exceptions to all rules...

Anchor diversity that I personally find solid:

Brand (name, domain name or url chunks) - 40%
Fluff (click here, get this, find out more) - 40%
Exact match (with variations) - 20%

You can go higher EM, but just how much higher is under review [several sites are using 40% EM and doing fine at the moment, but this could change].

Oh, and when I say exact match, I do mean a ton of exact match phrases and not just one or two keyword targets.

IMHO It is dangerous to set figures. We're seeing large variations especially in different niches. I'd suggest starting with 50/50 and adjusting depending on results. Do a competitor analysis and see what the current ranking sites have.

In the past year I've come to believe that a few social signals can have a big impact. So get out there and socialise... Don't need much, just a few mentions every month.

I'd also get rid of G Analytics and indeed any G tool that gives them extra information e.g. webmaster tools.

Look out for bounce rate and click through from SERPS and improve this. e.g. make your meta description compelling, use video to keep people on your site longer.

Stop worrying about dofollow and mix it up a lot more.

Look at getting traffic from other sources, forums, guest posting, social etc. (nothing to do with penguin really, but protects you from a total nose dive in SERPs)

Can't think of anything else just now. Our own sites are following the above rules (apart from social) and all have survived both Panda and Penguin unscathed.

 

[nitro23456]

Penguin isnt just about anchors.

Most of the people contacting me who have been hit by penguin have a whole host of backlinks coming from .co.in and .ph sources amongst other foreign TLDs.

You've got to ask yourself the question, if you're selling used cars in Bognor why the hell would you have 50 indian domains blogrolling you?

Google's answer is that you have paid some phillipino to spam the indian subcontinent for ten dollars.

Result = your site is bombing.

 

[Annoying Donkey]

Penguin isnt just about anchors.

I said mostly...

You've got to ask yourself the question, if you're selling used cars in Bognor why the hell would you have 50 indian domains blogrolling you?

If the same site had 50 Bognor sites blogrolling all with the same anchor text, then they almost certainly would be hit, depending on what other links they have.

On the other hand, if they had the 50 indian sites and 500 other non anchor text links, then they likely would not be hit.

It's quite conceivable that Indian sites would link to a used car site in Bognor, if they specialised in Morris Oxfords

 

[nitro23456]

what makes you think I was talking about what you said?

I was just contributing to the topic.

 

[Annoying Donkey]

And I was replying to what you said...

 

[nitro23456]

After initially quoting me as if I was referring to you.

 

[ethanarcher012]

Penguin is generally Over optimization and web spam penalty so simply take care of these 6 things and your website flourish.
1. Spammy Page Titles
2. Meta Keyword Tag Stuffing
3. Content Keyword Stuffing
4. Excessive or Unnatural Internal Linking
5. Reciprocal Link Pages
6. Excessive Footer links

 

[Motorcyclestore]

First reset every over optimized page, than resubmit sitemap in Google web master tool also add some unique or quality post..