Penguin 2.1 has caused quite a shakeup in the SEO world these last few weeks. Launched not long after Google’s Hummingbird update, there seems to be fewer people talking about this then there was for the introduction of Penguin or even its version predecessor, Penguin 2.0.
Though I previously shared that I had a lot of success on the back of helping people get over Penguin 2.0 – especially since I own Penguin2.com – I’ve yet to write anything about Penguin 2.1 until now. The reason is simple: I wanted to be absolutely sure I was giving the right advice on what to do after this update. After looking at hundreds of sites in my own network (yes, hundreds) I certainly am now.
One of the things that really struck me was how many SEO’s came out and said “my methods are working as well as ever; not one site of mine got hit by Penguin 2.1.” Needless to say, I’ve pretty much lost trust in a lot of those people. I’m actually proud to have had sites hit by the update; It means I’m testing enough strategies when it comes to SEO.
Announcing that not one single website you worked on got hit – especially when you describe yourself as an SEO expert – doesn’t give me a lot of faith in your practices. Unless of course you were lying, which really backfired in my opinion.
What I’m about to share with you is nothing mind-blowing. It’s not some secret algorithm loophole that finally got blocked in by Google. In fact, it’s exactly what I predicted it was on Twitter a few weeks ago in the first 48 hours after the update was made official by Google’s Matt Cutts.
You can see I quickly got to work trying to analyse where I thought the problem lied (with a lot of help from Diggy too):
It’s also what we’ve been preaching in the forums for weeks now, but I wanted to do a lot more testing before I ran a blog post about it. The only mistake I made in waiting so long to write this post is that the rank tracking tool that I previously relied on the most – SerpFox – doesn’t give you ranking reports older than one month.
I’ve since moved away from Serpfox to another tool I’ll do a review on in my next post, but it has ruined my plan of showing you some great screenshots. Basically I wanted to show you ‘here’s where a site got hit really bad and here’s our gradual recovery’. We definitely have the recoveries, but there’s no screenshot to show the hit and the recovery in one.
Though it may seem that these are being posted purely for egotistical reasons (wow, look how great I am at SEO) the truth is that I just want to show it would have been very easy for me to say I also have websites which didn’t get hit by Penguin. To say that my SEO is flawless and I didn’t lose a single ranking.
Of course, you could look at the dates and be skeptical about why I don’t have anything more recent. Damn, you got it me. Okay, here’s the real truth…
I could totally lie and say that everything we showed in Backlinks XXX and my own personal sites were not affected. That’s not true. In fact, we’ve added 8 exclusive videos to Backlinks XXX in recent weeks (yes, that’s 8) which go into Penguin 2.1 in a lot of detail and the tools you can use to get yourself out of any rankings drops.
The date lines up perfectly with Penguin 2.1, as do a large number of other hits we received. Keep in mind that we’re tracking hundreds of websites here and though a small percentage of them were hit, we did get hit. Unlike others, I don’t mind sharing that at all.
Shocking right? A guy who studies SEO like it’s as important as the future of humanity and yet I still have sites that got slapped by Google. But then there are all those dodgy companies who still send spam emails about doing directory submissions and guaranteeing number #1 rankings in Google so maybe we’re not the worst. Here’s the real, real truth…
And this team monitor hundreds of websites. This includes client of ours through our Private Inc. service (don’t waste your time going to the site, there’s nothing there), our websites which we actively try to rank and our sites which are nothing more than part of a high-PR network to build ourselves easy links. In other words, we have a lot of data to work with when this kind of thing happens.
Something interesting I want to share is that – for the most part – the sites which got hit and the sites which kept their rankings (or improved upon them) followed the same strategy.
The benefit of running our own network means it’s also much easier to dissect the sites which got hit because I’m handling all of the links myself. I can add and remove them all as I please. I’m not waiting for some disavow tool to kick into place from links I built elsewhere.
The following is examples of rankings which have increased (and thrived) since the Penguin 2.1 update
As I said earlier, unfortunately SerpFox doesn’t allow you to change the dates on how far back you can view a ranking (past 30 days) or my screenshots would look more impressive. Since I have (had) multiple SerpFox accounts, I never grouped the sites I was watching and therefore can’t change the date. I would have put up this blog post far closer to the event when there was actually buzz around it if I wanted to lie about this.
After editing these pictures to put on that fancy shadow I realised I had cut off the dates. Here’s one of the more impressive ones without the ‘special effects’ (I’m happy to share the rest in the comments):
Note that the rapid increase came a few days after Penguin was announced and has risen ever since. Diggy actually sent me a lot more than I included in this post:
Anyone who is a member of our private forums will know that what I’m about to share was figured out within 48 hours. The reason I didn’t jump the gun to make a blog post is that I had to make time to make sure our sites actually recovered and improved upon their rankings before I shared it here. It wasn’t so much that we had to find the safe, just wait a long time for the Redditor to come back and tell us what was inside it.
First of all, Penguin 2.1 only affected a percentage of search queries. This means a ranking may have dropped naturally without being hit by the Penguin update and we could suspect it got ‘slapped’ when it actually hadn’t. I know this sounds unlikely, but after looking at hundreds of sites I would say a couple do fit into this scenario. Even without our own data, it’s very hard to analyse other rankings in other industries because, once again, we don’t know if that was an affected query or includes affected sites.
There’s also the angle that different niches still have their own types of bias, whether they were hit by the changes or not. Some search results are dominated by freshness, medical terms are still dominated by old authority sites and some search phrases have such little competition that poor sites are still going to rank well.
Keep in mind that the original Penguin update, launched back in late April 2012, was clearly focused on removing spam from search engine results. I’m not talking about what it targeted here (thin sites, blog networks, etc) but rather the aim of Google in order to improve their algorithm.
While we’re on that topic, I don’t think there’s a specific type of website that got hit. In other words, it didn’t seem to matter too much where the links are coming from, rather that they fit into a certain mould of over-optimisation. I’ve tested this with Negative SEO on my own sites and found that to be true; though I’m sure there are exceptions.
For the last few years myself and Diggy have been operating under the idea that we don’t want our core anchor text to make up more than 30% of our links. To put this into simpler terms, if I’m trying to rank for ‘cat toys’ in Google UK and I have 100 backlinks to my site then I don’t want more than 30 of them with ‘cat toys’ as the anchor text. This is much higher than I’ve seen other people preach, but it has kept us in fairly good stead and gave us something to focus on.
Don’t forget that, once again, there are obviously exceptions. Google are a company worth around 250 billion dollars. They have the funds and the obvious motive to make a search engine that doesn’t play by the same rules across the board.
Now, our internal metric is much much lower than this. I’m focusing my main anchor metric on small sites to not be more than 10% of my links while Diggy is quite a bit more conservative at 5%. This is quite a drastic change from what we’ve been able to ‘get away with’ previously. Two good tools to check your current anchor text distribution are my constant recommended resources, Ahrefs and MajesticSEO. It’s worth creating a free account on each.
As many of you know, Diggy is my business partner when it comes to all of my SEO operations and he sent me a number of notes that he wanted me to cover in this post that he thought were relevant. On this second point, here’s what he had to say:
“Another thing that we believe the Penguin 2.1 update targets is the comparison of links to your inner pages vs the links to your root URL. Some people doing SEO seem to build a gazillion links to their homepages, but none to the inner pages. This is a clear signal of unnatural linkbuilding, and by increasing the spectrum of the update from just looking at the homepage link profile to the entire site’s links, it was once again easy to take out a huge amount of SEO’d spam sites that shouldn’t be ranking at the tops of the SERPS.”
When anchor-text diversification couldn’t really seem to be blamed on some sites that we monitored, having an ungodly amount of links to a site homepage and non to internal pages was what really set off some flags.
As I often preach, SEO is rarely more complex than trying to think of any algorithm updates from a Googlers’ point of view. If you were the head ofweb spam right now, what kind of things would you be looking for to find sites ‘gaming’ their way to the top of the system? This is yet another obvious tell that you should be more careful about unnatural link building on sites that are important to you.
I’m not telling you to get into pure Whitehat SEO (see point one near the end of this post) but it definitely helps to play into their logic and give your website a more natural link profile.
I’m probably starting to sound like a broken record here but I do not like sharing the results and tactics of other websites. However, without giving any examples at all it’s very hard to make points that backup what I’m saying to you. For the following two examples I’ve chosen very large companies both worth a few million, rather than highlighting some webmaster who relies on the internet income to save up for his dream car. I have those as well, but these two examples should be enough.
I wanted to show that not all sites avoid these two mistakes and have issues with ranking. Some clearly fly very much under the radar.
Clicky is an analytics software tool that I personally use and constantly recommend others to use via this blog as well. It’s totally free (up to a point) and you don’t have to worry about people tracking your websites via Spy tools like they can if you have multiple sites on a Google Analytics account.
Here’s some of the anchor text distribution for links that are pointing to the Clicky website:
As you can see, some of their terms have an abnormally high percentage of anchor text distribution. You would expect their website name to be right up there instead of their desired target keyphrases. Especially when they have millions of backlinks. Please note that this data was verified on other tools as well and while the amounts vary, terms like ‘Real time web analytics’ are found on a very high percentage of their backlinks.
So, how do you get millions of links for phrases like ‘Google Analytics Alternative’ instead of people just linking to your website URL like you would expect? Well, when you use Clicky tracking code, they sneak in a little linked image to show you use their service. The alt tag of the image and the title of the link both show the terms they want to rank for.
Fortunately it’s possible to remove this if you have even just a little knowledge of HTML (or find that very small checkbox in the corner showing how to turn it off) but it’s clear that millions of people haven’t done this. Every single time you refresh the page for your tracking code, a new phrase they want to rank for replaces the previous text…
I really wouldn’t have expected something like this to work in this day and age but it clearly hasn’t gone under any kind of manual review. I have no qualms against them putting the link in there by default – it grows the Clicky service and improves their offering – but the mixing up of those anchor texts is no doubt done to get top rankings for relevant phrases to their business. Something that is far from ‘natural’ surely in the eyes of search engines.
I took the hardest phrase they’re trying to rank for and you can see they’re sitting at a pretty 5th on my end:
Trust me when I say they’re doing much better for other terms you can find if you keep refreshing that tracking page code and see what they want to rank for.
Infogr.am is another example of a service which utilises its users in order to get better rankings for its website. Like Clicky, they have a service which means their end result is something you want to paste into the code of your website, which they really take advantage of. To be fair, Infogr.am seem to offer a great service and have built up a brand at a great time where Infographics still seem to be all the rage is a digestible way to enjoy content.
They’ve also been fortunate enough to get regular embeds on one of the biggest blogs in the world, The Verge. Here’s one of the more recent graphs posted on The Verge courtesy of Infogram:
If the arrow doesn’t give it away, that’s a nice juicy link at the end of every embedded infographic. This time I will show you an anchor text distribution chart from Majestic SEO just to keep things fair.
As you can see, the majority of their links are for the text ‘infographic’ (how lucky!) and over 18% of them that have been picked up by this tool are for ‘create infographics’ which can be found embedded in said infographics. Even luckier is that it does transform into rankings as well:
In both cases here, just by looking at the source code of the links, I would not have thought of them to have a big impact on search rankings. After all, the links for Clicky are just image links with some ‘alt and ‘title’ info and while they do help, it’s still an image link. Similarly, they are linking to affiliate links in that code so they must be doing some sort of 301 redirect from the affiliate link back to their homepage in order to pick up that link juice.
If you look at the source code of Infogr.am links on those charts then you’ll also see that the link isn’t very ‘clean’ when it comes to HTML standards. They have the link code in there with the desired anchor text, but it does look a little messy. It’s still clearly working though.
This is the kind of thing that Google have publicly said they don’t want companies doing over and over again yet it still seems to be working. My main point here however is that just because I recommend your anchor text for your ‘money terms’ to be under a certain number, it doesn’t always have to be. Clearly.
The most obvious lesson first of all is to head into Google Webmaster tools and see if you have any warnings about your site. If you have been a ‘bad boy’ and building bad links, you may need to look into disavowing some of them to have any chance of a comeback.
#1 Make sure you have a website which breaks all of the rules
This major update, and big updates prior, have not pushed me to completely clean up my act. I still think it’s very, very valuable to have sites which go against the conventional methods of building a website that Google will love. It’s very easy to spend $50 on Fiverr and run a few thousand dirty links to a test website or pick up a copy of GSA (full video guide inside Backlinks XXX) and run that on a VPS 24/7. Keep the website thin – just a few pages – and use an expired domain if you can to give yourself a little credibility.
Not only do I think you’ll be surprised that these kind of sites can have quite a bit of life in the right industry but you’ll also be able to figure out what updates hit for yourself without waiting for other people to share their findings. This is especially the case if you have a few dozen websites or, like us, a few hundred.
#2 Slow down on what you’re doing to have an unnatural profile
Hopefully one or two of these points fit in with your own site if you have been affected by Penguin 2.1. The simple idea then is to ease up on what you’re doing for this to be the case. If you have an overly high anchor text distribution for certain words and phrases then consider changing those up in continued link building efforts.
Similarly, if you’ve been solely focused on building links to your homepage or just a small percentage of your site then slow down those operations too. There are cases where Penguin 2.1 (and all of the Penguin updates) seem to have hit some sites unfairly, but in most cases there are a lot of unnatural elements to a webmasters SEO efforts. If that’s the case for you, then you need to be focusing your efforts on other avenues.
#3 Diversify your anchors and internal linking
I’m not going to turn point-three into some huge section on how to build backlinks. I’ve covered that in a huge blog post in the past which is still just as relevant today as it was back then, bar one or two examples. Something to keep in mind is that your links don’t have to be followed links in other to help with your anchor distribution and getting links to internal pages.
Meaning, you don’t have to get high-quality links in order to help with your backlink profile. Simply getting links from places like blog comments, forum signatures and guest blogging (again, even with nofollowed links) should help you to build a more natural, diverse set of links.
Since it’s soon to be Thanksgiving in America, the internet is going a little crazy with all of their Black Friday deals. I don’t like posting on Friday’s since less people are online, so I’m making this offer available now. For the next 48 hours, you’ll be able to reserve a 100% free SEO call with myself or Diggy to discuss your website for 20 minutes.
We’ll get on Skype with you and look at your website and offer advice in real time and you’ll be able to ask us any questions you have as well. Since we wouldn’t be able to handle hundreds or thousands of people taking up this offer there will be a $500 deposit required. This will be 100% refunded after the call, Paypal fees and all.
The deposit is simply to make sure that people who reserve a slot to chat with us will actually turn up for the call and we’re not sitting in front of Skype waiting when we could have been doing something more productive. If you aren’t able to do this in the next week then I would refrain from reserving a time.
To guarantee your spot, here’s our link to Paypal for payment. Sorry, received more than I expected so have to close this early.
I tend to do this kind of thing once per year to have a fresh idea on what problems are bothering people (last year it was Blogging focused) and what common mistakes I see people making with their sites, so we hope to get as much out of this as you will with the free session.
Have a great Thanksgiving to all of our American readers and to everyone, thank you for reading as always. If you’ve had a site affected by Penguin 2.1 then I would love to hear your story in the comments…