Penguin Diagnosis and Solutions for Penalised Sites
Last month was a bit of a horror show for a number of people. The Penguin update killed a lot of Webmaster’s sites (far more then Panda did according recent SEOMoz analysis) and now everyone is looking for the answers.
The annoying thing about this update is that everyone has tried to explain the synopsis and what went wrong.
I find it really irritating when major SEO blogs such as SearchEngineLand.com and SEOMoz.org write up posts on “Penguin Diagnosis” and “How to Fix your Site” etc when they don’t have any solutions or knowledge of what is going on. It’s like they feel compelled to write about it, even though they don’t know what to say. “Don’t panic” seems to be the main advice, which isn’t exactly advice in my opinion when people have been losing their jobs.
My Own Penguin Recovery Advice and Post-Penguin SEO Tips
First of all, the Penguin penalties for most webmasters were because of poor quality links. People are arguing about anchor text, analyzing why specific pages moved up or down, but the fact is this Penguin update just penalized web spam with tons of low quality links.
People keep going on about how it’s “anchor text optimization”, “on-page optimization”, “a penalty against affiliate sites”, blah blah. At the end of the day most sites that have bought spammy links all fit these profiles, which is why there’s so much correlation. I think a few other factors might have been involved (e.g. brand signals increasing in importance, domain age), but in general this update was all about targeting web spam, not promoting high quality sites.
Just to end the debate, my own affiliate blog was hit by this update. I have 100% original content, a range of loyal followers (including daily type-ins), my brand name AffiliateFYI.com in all of my Page Titles, tons of excellent, natural links, excellent user metrics (average 5 mins on site, 30% bounce) etc…So you can’t say any of that was the issue.
So what did I do wrong? I bought a ton of shitty links from DigitalForums over the last few months, which I now regret. Regardless of whether you have a good site, Google has penalized any sites that have bought poor quality links, which is outside of their Guidelines. Top sites have gotten away with it for obvious reasons (i.e. they are good enough for Google to ignore these things).
Advice for Webmasters Affected by Penguin
My personal advice is don’t wait around like all the SEO blogs suggest. They don’t really know what they’re talking about and just aren’t confident to give advice on this update.
1) Since this update targeted low quality links, you need to work out if you can remove those low quality links you bought. If you can, great, and I’d hope that this rolling algorithm would de-penalize your site after a month or two (I don’t think there will be any re-inclusion request).
2) The chances are you can’t, since most low quality links will be from automated SEO programs, directories, forum profiles, auto-generated web 2.0 blogs, and webmasters with no contact address. If this is the case, I’d personally take your site down, wait for it to de-index, the re-launch the entire site on a new domain (and switch the quality links to the new domain). This might not seem a perfect solution, but it will get you back on track and remove the risks of future updates that target web spam.
Is it Worth doing a 301 for a Penalised Site?
301’s tend to work for penalized sites, for a while anyway. I’m currently doing an experiment with one of my penalized sites and so far the 301 has worked and restored rankings.
The problem with performing a 301 is that it’s a short-term solution, like a ticking bomb. It might re-rank your penalized site for a month or two, but I think eventually the penalty will pass through and you’ll be back in the same position. Hence, if you have old sites you don’t really care about then I think it’s definitely worth considering doing a 301. On the other hand, if you have a site that you invested a lot of effort and content into then you’re better off starting on a new domain with a fresh start.
In terms of planning new sites for the future, I would now focus 100% on brands (.com/.co.uk sites) with lots of excellent content, social media following and coding. Yes, links will still be important, but you’ll need to rely on high quality links from good sites, rather then low quality links from spammy sites.
Going forward, change your expenditure from links into great content and social media, since these are risk-free assets that Google cannot punish you for. Lots of content will lead to lots of traffic. Even if your site gets penalized you can recycle great content so it’s good to have most of your value in it. Penalized exact-match domains on the other hand will just become garbage. Top EMD domains are still valuable in terms of SEO, but I just think the risk in buying them has increased, plus it’s less risky to develop a brand for free in my opinion.
How Trustworthy are Social Signals as a Ranking Algorithm in SEO?
Yet again, I’ve been thinking about the future of SEO and the effectiveness of social signals as a ranking algorithm.
I still can’t see how social signals can provide any kind of effective, ranking algorithm for search engines to use.
Ignore the fact that social signals will be ten times more susceptible to gaming then links in 12 months time (they are cheaper and quicker to obtain), the types of things that friends will “share” or “like” with their friends are completely different to the types of pages that will get linked to.
The Difference Between Socially Shared Content and Link Content
Websites and authors tend to link to website pages which are relevant or which they reference in an article.
Links tend to be timeless and they accrue naturally over time. A well-written bio page on “Gandhi” or “how to backup your wordpress site” will get linked to on a regular basis from other relevant sites, allowing Google to see what is popular and authoritative.
The types of things that get social signals on the other hand tend to be ground breaking news, gossip stories and other interesting things across the web. As such, the majority of social signals are extremely short term and “viral” in nature.
Given the social element of liking, tweeting or +1 content, visitors are also going to limit to type of pages they share. Anything private or medical content is not unlikely to be shared, nor is anything that isn’t ground breaking.
I might really enjoy reading a 4-year-old page on Texas Holdem Odds or the life and career of William Shakespeare, but I’m much less likely to share it with my friends. Overall, social signals tend to be used to highlight groundbreaking news, gossip stories, pictures or the serendipity of browsing the Internet. There are certain things that are just aren’t built for sharing. A “like” isn’t the same as a personal link, instead it’s someone saying “hey come check this out” to their friends.
Social signals are too “spiky” as well. A news story will receive 1,000 likes, then disappear to the abyss of the archives. How is Google meant to work out whether this story is still relevant or up to date?
When a website links to another website however, that is a timeless piece of reference passing authority and web credence to another site. Website links have much more longevity then a human “like”, and most sites don’t link out willy nilly. It’s still an editor-conscious decision, and it’s a system which helps bring order to the Internet. It allows search engines to rank websites, understand the nature of their content, relevancy, establish a hierarchy of website structure and internal pages, whether someone is still popular, and so on…
Social signals provide no anchor text, geo-targeting or relevancy signal for Google. Just because one person on the other-side of the world likes something doesn’t mean someone else will (Just like an english .co.uk website might link to something that an Israeli .co.il site wouldn’t). If a 5-year old likes a cartoon it doesn’t mean a 50 year old professor will either. So, it’s really hard to any search engine to tear anything apart from these signals, unless of course they know the person, his profession, his age, and can recommend that sort of stuff to others in his circles (which is essentially how Facebook’s advertising is run).
This is why Google will be investing so much money in G+ and getting users on it (TV Adverts and encourage people to use it). Social signals are pointless unless you can analyse who is sharing what and who they are, what they like. When a website gets a link from another site, google analyses the relevance of the website where the link originated Why shouldn’t this be the same as social signals? I still can’t see G+ rivaling Facebook however because there is no incentive for people to migrate from Facebook to it. It’s a ghost town.
That’s one of the many reasons why I can’t see Google Authorship becoming that important. The other day I clicked on an author from a random Guardian article in the SERPs and the author had a completely, bone-dry empty G+ account. What’s the point in implementing a rel=author account if you’re not even actively using G+, lol.
In conclusion, while I think social signals will still be important in the overall SEO algorithm and in distinguishing between good and bad sites, they can’t possibly replace the effectiveness of off-page link ranking signals over the last 15 years. At least until there’s a bigger breakthrough in the social networking side of things with regards to the semantic web.
March Affiliate Report and April 2012 Goals
I know it’s not the end of March yet but I’m going away on holiday for the next 2 weeks and wanted to get this recap posted now.
Overall, it was a bit of a disappointing month with regards to business. Income-wise it was fine but a couple of my trading sites are taking longer to generate traffic and revenue then I expected.
This isn’t that bad however, since I’m just building them at a slow pace and not doing anything silly with SEO. Slowly but surely I’ll see the results, especially with great content and information. I’ve also more then doubled my Adsense revenue on my finance sites then compared to February.
The only new domain I bought in March was MobileRecycling.net, which will make a very nice mobile recycling comparison site. I’m already in talks with a couple operators about developing a white label on my site, which would be really cool. I hoped to conclude a deal for another new domain this month but I’m still waiting on that. I’ll keep you updated if that ever happens, lol.
My mobile gaming sites have been doing pretty well. I 301’d one of my penalized domains to a new domain and it’s performing much better, plus I’ve stuck up a page on my new OddsWinner.com sports betting site help it age and get indexed.
AffiliateFYI.com Income Update
I didn’t do much SEO on my blog this month, so traffic numbers didn’t increase. I’ve made around $100 this month (on top of the $100 last month), which is fine, but I’m starting to struggle with ideas about how to really monetize the site.
On the one hand, I’m getting a really good CPC with Adsense, but I think a monetization strategy that involves Adsense would be wasteful on this blog. I’m sure there are plenty of better ways to monetize the site when you’re providing a guide for new affiliates. As always, I’m focusing on quality information and building traffic first, and the monetization method can come afterwards.
I’ve also done some more guest posts on some very high traffic affiliate blogs, which have helped me pick up a bit of traffic.
I added my email capture and free SEO ebook that I’m giving away, although I’m seeing MUCH lower conversions then I anticipated. It’s made me more skeptical about how well my premium ebook will convert if I can’t even get people to download a free ebook! Having said that, my aim of getting 300-500/day traffic should be completely different to the 50-70/day I’m getting right now.
April Affiliate Goals
Not much to say here since I’m going away for the first 2 weeks. My main aim is to pick up some CPAs on my Forex and Binary Options sites and possibly even sell a couple of sites in my network. Other then that, it would be nice to conclude a deal for the domain I’m after and earn a bit more money from this blog!
General SEO Update
I couldn’t write a monthly affiliate report without talking about the changes that with regards to Google de-indexing many of the major blog networks and submitting unnatural link building warnings to webmasters
Truth be told, I didn’t receive one of these, but it has changed my SEO strategy going forward.
I’m only focusing on building high quality links from now on with very diversified anchor text and a lot of brand building. I still think these changes have increased the value of premium domains, but things like this definitely bring you back down to Earth. I’ve seen some huge sites that must have been earning at least $xx,xxx per month lose their rankings, so it pays to be more smart then agressive.
Trying to Put my Finger on the Future of Google and SEO
I’m rambling out loud here, but I can’t help but think of stuff like the future of search and “what could effect my business” type of scenarios.
Let’s cut to the chance…
Factors that could Influence Google and SEO; 2012-2015
First of all, let’s look at where Google’s priorities lie: improve results for users and reduce spam in the search results.
Obviously, the trouble for Google is that while there will be a variety of signals that they can use to identify authoritative, relevant results for the user (e.g. links, social metrics, keyword content), people will always try to game the system.
Everyone knows that social will become the next big thing in SEO – just look at the launch and integration of Google Plus in the Google.com SERPs, or how Twitter and Facebook Shares can help SERP rankings – especially in Bing.
However, just like links, social can be gamed in one way or another. Buying real Facebook followers or +1s will always be a possibility (just look at SocialADR). Not only that, but if social signals do become more important then the number of services supplying social signals will increase, and get better, making it easier for webmaster to chase the money.
Moving from Binary Ranking Signals to a Continuum of Signals
The Panda update in February 2011 did more then just evaluates on-page factors such as content and ads. It allowed Google to draw the line between spam results and high quality results, based on mathematical algorithms for finding high quality sites.
What we have to remember however is that Panda is just a filter or a penalty. That is, it creates a binary format of plugging good results on one side, then bad results on the other.
I’m not sure exactly how effective Panda would be as a ranking algorithm in itself. That’s probably what Google’s engineers are working on right now. Because, when it comes down to it, Panda was about trimming the poor results out of the index as a direct result of the Vince update. It wasn’t designed to be a ranking algorithm to replace links. Yes, it helps find good sites, but it doesn’t help rank pages or find relevant results for users. That’s where links and anchor text become really important.
Why Links are so Important to Google
Links aren’t just a matter of identifying authoritative brands. They help Google identify the most important pages on the net, what the content is about, when a page is credited, what is popular when, and other relevancy metrics.
The main problem with brand metrics if you’re Google is that yes, brand queries, citations and user engagement show what a good site is, but they don’t help prioritize the importance of internal pages or figure what they’re about.
One of the main thoughts that popped into my head is that Google’s robots would become a lot smarter in understanding how a brand is associated with a certain topic through semantic proximity and vectors.
For example, if the brand “City Index” is widely used next to the words “spread betting company” or “spread betting”, then it would help City Index to rank for that term. If you want proof that Google is capable of doing this then just search “Forex Brokers’ or “Poker Sites” in Google.com. It displays a list of relevant results at the bottom of the page such as “Full Tilt Poker”, “Party Poker”, “Carbon Poker” or whatever, because these are the most cited mentions next to the words “Poker Sites” on the Internet.
Like I just said, even if Google did this, it doesn’t help prioritize or find the most important pages. For this, we need links (or arguably social metrics).
The Role of Social Media in SEO
Social metrics will obviously play a role, but just like links, they can be gamed relatively easily, and unfortunately they don’t provide the “anchor text” relevancy that Google needs for important pages. Social metrics let Google see what is high quality and popular, but not what the page is about.
This is why I think Google will have such a hard time moving away from a key reliance on links.
If we reverse engineer everything and go back to basics, we need to ask ourselves “what is a high quality result for the end user”. And what it mainly comes down to is answering the searcher’s question. Most people who use Google are just looking for information, such as the latest football scores or a comparison of ISA interest rates. Whether we go above or beyond our user’s question doesn’t matter that much in the face of things – we either an answer or we don’t, and the user leaves. Obviously we want the user to stay on our site and come back later, but in terms of answering the person’s initial query we’ve done our job. Think about it – we don’t use Google to find brands (we already know popular brands), we use Google to find things, products or information.
That’s the other binary issue for Google. A search result is either good enough or it isn’t, which is why off-page factors such as links and social metrics will always be important for ranking. On-page factors probably don’t provide enough key signals for ranking the entire Internet!
Rambling about Spam vs. Google vs. SEO Advice
I think the whole SEO industry is pretty messed up, for several reasons.
First of all you have Google, who wants to rank the best results and actively encourage webmasters to SEO their sites, hence why they have made over 400 “how to” videos on SEO for webmasters.
What this does however is educate webmasters about SEO. And then they realize that they need to buy links to rank their sites above their competitors in Google. It’s not just low quality spammers doing this – everyone is, including the biggest brands such as Barclays, BMW, Santander, Virgin and yes, even Google.
But then you have Google, who can’t exactly give the OK to webmaster to build links (it would completely destroy their ranking algorithms, it would make them hypocritical if they decide to move away from link signals, they would lose a lot of their biggest Adwords clients to SEO companies and drive Adwords revenue down), but at the same time they have to keep face and maintain a working relationship with webmasters.
Whilst this is all going on, Google is trying to stop spammers by penalizing them based on over-optimization – but there’s clearly a limit to the extent they can get away with this because a) their algorithm is still dependant on link authority and anchor-text relevancy and b) many of the biggest brands have engaged in blackhat or greyhat tactics.
The final issue is that whatever algorithm Google comes up with, people will constantly try to game it. Search engine traffic is far to valuable for anyone to give up on, so essentially Google is constantly going to be up against it. At the end of the day, it’s possible that they could just say, “Look, we acknowledge that that brands are buying links, it’s a free competitive market, there’s not much we can do about that, all we can really do is remove spammy results from our index.”
Most Innovative Ranking Solution for Google to Use
Rather then becoming over reliant on certain signals, Google could compare and contrast various signals on a site and make them all balance each other out.
So for example, let’s say you have 8 results on page one with excellent content and user engagement, but then two results with OK content and engagement, but a ton of links which have pushed them up there. Google could then say “hmmm, this guy isn’t spam, but we should try to devalue his over optimized link profile so that it correlates with his content better and so he fits in with everyone else”.
In other words, a bad score on one part of Google’s checklist could bring down the other parts, and visa versa. So if you do have great content, then Google could then give a site with a weak backlink profile a bit of a handicap or boost. If you have poor content, a low volume of brand query searches or poor user engagement relative to the other sites, then your backlink profile could be devalued slightly.
Essentially, it’s about preventing any one signal becoming overly important, making it extremely hard to spam the results. Individual signals such as links would stop becoming a fixed importance in the ranking algorithm (e.g. 60%) and instead they would play off each other and vary between sites. That being said, these are just anti-spam measures and they would not stop everyone from still buying links.
The Future of White Hat SEO and Downgrade of Blog Networks
It’s been a hectic last couple of weeks for blog networks and the SEO industry. I just want to add my two cents on what’s happened and on the future of SEO.
2012 Signals the End of Blog Networks and Black Hat SEO
I think the last 2 weeks have signaled the end for black hat SEO completely, which includes blog comments, link pyramid schemes, web 2.0 blogs, blog post networks, etc…
A lot of the major blog networks have closed down or had the majority of their blogs de-indexed (e.g. BMR, ALN, the Link Juice Network) and I’m sure more of these networks are to come.
Since the introduction of the Panda update in February 2011, it seemed obvious to me that black-hat SEO was going to die soon. The SEO strategy of spinning hundreds of articles for blog post networks or engaging in link pyramids with different “tiers” was always going to become obsolete. Everyone thought that they were protecting themselves with these “tiers” (you can read about these in the SEO eBook I’m giving away on my blog) but the truth is it’s fairly easy for Google to follow the breadcrumbs. Once the base of your black hat pyramid scheme (e.g. forum profiles and spun articles) become de-indexed, the knock on effects will be massive and so this was inevitable.
I think the smartest black hat SEOs knew that this would happen eventually, but I think a lot of less intelligent black-hat people became over reliant on these strategies without considering the risks. Either way, I guess we’re all still surprised at how quickly Google has acted on this. I personally thought the majority of blackhat SEO methods would work for at least another year until Google caught up.
I’ve already seen a lot of people in the black hat SEO forums squandering about what to do now, which blog networks they should use, how to replace their backlink strategy etc, so here’s my opinion…
My Opinion on the Future of SEO
Blackhat SEO is going to be out the window. Forget that, it’s a lost cause. There are far too many negative signals that Google has caught onto for this to work including: low quality content, low quality blogs, lack of social signals, low quality links, over optimized backlinks etc…
I think even grey hat SEO methods from last year have become outdated. The biggest issue of all is that you can’t escape from the past. Even if you engaged in black-hat link schemes on your website from years ago, and have cleaned up your act now, the algorithms can still hurt you and penalize you based on historic strategies.
I think the future of SEO right now literally revolves around building a high quality site for users (good, fresh content, social metrics and user metrics) and diversified, high quality backlinks and guest posts. Over optimization looks like it’s going to be a thing of the past. I think going forward I’m going to focus on natural brand URLs and just try to get some general authority going with my sites. Oddly enough I think these changes have also increased the value of premium exact-match domains.
The fact remains that SEO isn’t going to die. Google’s algorithms still rely heavily on backlinks and anchor text for relevancy signals and authority. This means that a good backlink now will always be valuable, regardless of what the critics say. Nothing will ever replace links in the search algorithms – just look at the past 15 years of the industry.
Links are still as good an indicator of relevancy and trust now as they were 10 years ago. Google just has to find a better way of devaluing the lower quality links from poor sites with no traffic. De-indexing half of the blog networks was a great start. These really were the lowest of the low in terms of web content and I don’t think too many people can complain. I don’t want to sound too callous, but these changes will probably eliminate around 25% of the competition in the search engines. I’ve already seen quite a few competitors with poor sites dropping places.
I’ve also been spending a lot of time researching the major black hat SEO forums and some of these people have been completely dependant on BMR for their rankings, with 2,000 links to their site from BMR. Thus, it’s hard to feel sorry for some of them. Changing to the grey-hat or white-hat method of SEO is going to become a lot more expensive or even impossible for some of them, but I think Google had to do something about blackhat SEO and automated software sooner or later.
One last point I want to make is the people arguing that Google should devalue spammy links rather then penalise webmasters, otherwise it would be too easy to link-bomb your competition. I think this is the wrong way of looking at things however. What Google has been trying to accomplish with these unnatural link building penalties in my opinion is prevent people optimising their sites for specific keywords. If you have 100 links to your site, and 60 of them have a targeted keyword unrelated to your site, Google isn’t going to be capable of knowing which links to devalue. They can only force you to remove these links manually and stop trying to optimise yourself for keywords.
Recent Posts
About Me
Hi my name is Adam and this is my affiliate blog. I've been working in the online marketing business for two years and graduated from Uni in July 2010.