Fighting Penguin: A True Story

9 comments

One of my clients got smoked by Google’s Penguin update.  We plummeted from the top spot for multiple target keywords to out of the top 100 results.  This post reviews why we were penalized and how we plan to retake our position.

How the Mighty Have Fallen

My client ezdiquote provides online disability insurance quotes.  Our target keywords include long term disability insurance, disability insurance quotes, and income protection insurance.  The following table shows our spot-checked rankings taken at the end of each month.  Notice the precipitous drop at the end of April following the Penguin update.

I am sure many of you witnessed it, but the change was literally overnight.  April 23rd we had solid rankings and steady high quality traffic.  April 24th we were invisible for our target keywords and essentially cut off from organic traffic.

Why Google Struck Us Down

As the parameters behind the Penguin update become clearer, the reasons why ezdiquote lost its rankings begin to make sense:

1. Site-Wide Links

We used site-wide links with commercial anchor text from our microsites.  We have multiple microsites targeting specific occupations (e.g., itdiquote.com for IT professionals).  We use the microsites for LinkedIn and Facebook advertising where we demographically target by occupation.  On each of these sites, we included a link in the top navigation to ezdiquote.com using target keywords (e.g., disability insurance quote, long term disability insurance, etc.).

Figure 1: Site-Wide Commercial Anchor Text in Top Navigation

Admittedly, we chose to use exact match anchor text precisely because we could.  We controlled the sites; why not link with our target keywords?  Unfortunately, this is exactly what Google targeted with Penguin – site-wide links with exact match anchor text for competitive keywords.

2. We Used Private Blog Networks

Like many SEOs, I loved private blog networks.  Where else can you quickly and easily submit content and get it immediately picked up?  From my standpoint, we provided unique content (we never spun articles) that benefitted the web.  The content linked to us, but it was also (at least somewhat) informative.

3. Poor Anchor Text Distribution

With private blog networks, we intermixed our target keywords with our URL, brand, and full sentences to mimic a more natural link profile.  It simply wasn’t enough.  Coupled with our exact match links from our microsites, our anchor text was heavily skewed to our target keywords.  Natural link profiles typically have 20% commercial anchor text.  We had closer to 60%.  Again, we were easy pickings for Penguin.

The Road to Redemption

Our problem is clear.  Our anchor text is too densely skewed to our target keywords.  Pretty easy fix pre-Penguin, but I don’t think merely adding non-commercial anchor text links is going to salvage our rankings.  Our strategy needs to be multi-pronged:

1. PR is Back

While Penguin was detrimental for many SEO strategies, it was a godsend for PR agencies.  While anchor text used to be key, sharing and respectability now claim the throne.  SEOs need to get our clients’ stories out there – whatever story that may be – and we either need to rely on PR professionals or gain some PR chops ourselves.

For transparency sake, as you may have surmised, this blog post is a part of that first step.  In the SEO community, Penguin is newsworthy, and so I have a great opportunity to share my client’s story and gain some press.  To be truly effective, we need to do this 100 times over, changing our story to make it relevant to whatever audience we are pitching.

2. Learn from Competitors

I have repeated this step dozens of time for ezdiquote, but it is an especially important step after major algorithm updates.  We need to look at what sites benefitted from the update.  Their link profiles provide insight into what strategies continue to work.

The site that ranks best post-Penguin has a very similar link profile to ezdiquote.  The site has multiple inbound links from microsites it owns plus some low quality article and directory links.  The main difference is that the competitor did not use exact match anchor text from its microsites.  Rather than “disability insurance”, the competitor used “Guardian disability insurance” in its anchor text and alt image text – an incredibly subtle difference, but in Google’s eyes “Guardian disability insurance” is a low volume, non-commercial keyword, whereas, “disability insurance” is an incredibly competitive, high volume keyword.

Based on this insight, we have replaced our exact match anchor text on our microsites with image links with alt text: “ezdiquote disability insurance quotes”.

Figure 2: Updated Links on Microsites

3. Rebuild the Foundation

Lastly, we plan to focus on foundation links and citations such as relevant directories and partners.  These links include human-edited directories such as Business.com and Best of the Web.  Whereas, when we initially focused on building foundation links, we looked at including target keywords, going forward we are largely looking to diversify our keyword anchor text with low volume non-competitive search terms.

Part of building foundation links is building links to our backlinks.  If Google is unable to index our backlinks, the backlinks are useless.  We will submit content-focused backlinks to social media sites (e.g. Delicious and Reddit) or link to from Twitter and ping search engines with our back link URLs via services such as Ping-O-Matic.  This should help keep our backlinks in Google’s index.

Putting It All Together

The immediate goal of our post-Penguin refresh is to establish a more natural link profile without losing the benefits of our previous (and previously successful) link efforts.  New links from PR, ezdiquote microsites, links “borrowed” from competitor link profiles, and foundation links with anchor text focused on low search volume, non-competitive keywords combined with our existing exact match anchor text links will hopefully help us to return to our once dominant position and prepare us for the next algorithm change (puppy update?).

About the Author

Mike has worked in the online marketing industry since 2005. Mike started Obility Consulting in 2011. Obility works with b2b companies looking to lower customer acquisition costs through improved performance in paid search and conversion optimization. Please visit obilityconsulting.com to learn more about Mike or Obility.

Add Your Comments

  • (will not be published)

9 Comments

  1. I would recommend using tools like iBacklinkPRO or MajesticSEO to research your competitor's backlinks they can help a lot with getting natural backlinks

  2. Fred Waters

    From my experience in my niche almost all of my competitor's links are not natural. Some are better at getting unnatural links than others. Also, this update still needs some work. I've seen some of the worse sites with the most obvious spam links rise to the top.

  3. Great post.. Ive done the same thing with mciro sites & blogs! Its to tempting to not use exact keyword anchor text when you control the site... Im def guilty and took the pounding! The cool think is I can change it haha! Thanks again!

  4. That sounds like a good strategy you've chosen to adopt. A lot of people have made the mistake of either not diversifying their anchor text enough or focussing on just commercial oriented terms. I've personally seen people who've been hit by the Penguin actually use directory submissions to help skew the anchor text profile back in their favor and which has worked for some - e.g. getting links with the anchor text as "EZDIquote.com", "EZDI Quote", "EZDI Insurance", etc + this has worked better when the official site name is also mentioned on the title of the homepage/other pages. Right now the title of the homepage doesn't have the brand name, perhaps it could be mentioned at the end of the title. That would make such links obtained look more natural and to be of more sense. Just my thought.

  5. Good article! Google updates make me nervous.

  6. Excellent article. Solid advice there in examining the new top of the SERPs for the new keyword density percentages. I'll be using your article to help poorteds move up in the world.

  7. I can't help but wonder how many of your micro sites were linking to your homepage. It might not have been the consistent text you were using as much as it was that you had the same text all going to one single page of your client's site. Just a thought.

  8. Have you already seen any signs of recovery?

  9. Thanks Mike for an honest post. I'm so tired of reading blog articles that say "I'm proud to say none of my sites have been hit by Panda/Penguin and here's why..." and then say what "old-tactic SEO's" do and they would never do such a thing. Honesty is what we need. Great post.