This case study was created using an LRT Superhero account.
Some of the use cases explained in this case study are not available in lower plans.
The LRT Superhero Plan (and higher) includes all our 24 link data sources and allows you to perform link risk management, competitive research, professional SEO and backlink analysis for your own or your competitor's sites. You get to see your website's full backlink profile picture and this can make all the difference for your SEO success.
A Case Study about a Penguin 2.0 Loser
and Step by Step Guide to Protecting Your Site
Let’s take a dive deep into the reasons for the drop in visibility that elearning.com experienced after the global rollout of the Google Penguin 2.0 update.
This research looks into Quick Comparisons, Competitive Analysis, Anchor Text Analysis, Link Detox Risk Analysis, and much more.
This is our 4th deep dive case study. A lot of information and details are presented here, and we’re stoked to deliver you the next case study by a Certified LinkResearchTools Professional.
We would greatly appreciate you sharing this quality piece of work by Marcela De Vivo.
Christoph C. Cemper
How we will look into this Penguin 2.0 Penalty
At first, I will have a look at the loss in visibility on a root domain, sub-domain and keyword level. Following to that I will identify those areas mostly affected in order to derive the hypotheses for the detailed analysis
- Quick Domain Compare (QDC)
- Quick Competitive Overview (CLA)
- Link Stats Comparison (Juice Tool)
- How do various link factors compare?
- Summary of Findings from Quick Comparisons
- Link Status (CLA)
- Deep Links Ratio (CLA)
- Sitewide Links Ratio (CLA)
- Referring Class C (CLA)
- Moz Domain Authority (CLA)
- Google Page Rank (CLA)
- Link Velocity Trends (CLA)
- By Retweets (CLA)
- By Google +1's (CLA)
- TitleRank Home Page (CLA)
- Link profile By PR & AC Rank (BLP)
- Summary of Detailed Competitive Analysis
- Keyword Breakdown Money vs. Brand links (CLA)
- Money Keywords (CLA)
- Anchor Text Distribution compared (QBL)
- What is the Anchor Text Distribution for Devry (QBL)
- What is the Anchor Text Distribution for Kaplan (QBL)
- What is the Money Keyword Distribution for Kaplan (BLP)
- What is the Anchor Text Distribution for Phoenix (BLP)
- What is the Anchor Text Distribution for Capella (QBL)
- Summary of Findings of Anchor Text Analyis
- Link Detox Overview (LD)
- What is the Average Link Detox Risk?
- Do any of the links stand out?
- Scan Combined Backlinks (CLA)
- Network alert! Network alert!
- Identical sites on different domains
- Spot Test Elearners Backlinks (BLP)
- Paid contextual links
- Paid Links on USAToday?!
- “Partner Links” on Studyabroad
- So what gave Elearners away, and caused the Penguin 2.0 penalty?
- So why did Elearners.com get a Penguin 2.0 penalty?
- Too Many Unnatural Ratios
- Too many links with Money Terms in their Anchor Text
- Too Many Paid Links
- Elearners is part of a Link Network
- So what does this Penalty tell us about Penguin 2.0?
- Watch your Ratios!
- Watch the number of Money Terms in your Anchor Text
- Use Brand and Noise Terms in your Anchor Text
- Don't Buy Links!
- Avoid Link Networks
CEMPER Power*Trust is now LRT PowerTrust
You may still see CEMPER Power*Trust™, CEMPER Power™ and CEMPER Trust™ on some screenshots in this case study.
In 2015, we renamed these metrics to LRT Power*Trust, LRT Power and LRT Trust to reflect the shortname of LinkResearchTools - which is LRT.
Penguin 2.0 hit hard for those who didn’t know how to protect their site from penalties. Elearners.com was one of those sites.
According to SEOlytics, in the aftermath of Penguin 2.0, Elearners lost close to 60% of their traffic.
For all intents and purposes, Eleaners.com should have NOT been hit by a penalty when Google updated to Penguin 2.0. Taking a superficial look at their metrics, they had all the right elements: high PR/authority links with a large number of unique C class links, including .edu and .gov sites. So why did they suffer a Penguin penalty? And what can you learn from their mistakes to protect your own site?
Penguin penalties are preventable — if you know what you’re looking for and how to protect yourself.
By using LinkResearchTools.com, I’ve undertaken a deep, step-by-step analysis into various backlink metrics, revealing numerous red flags that—seen from Google’s perspective—created unnatural ratios resulting in a harsh penalty.
In this case study, I will lay out a step by step strategy that you can follow to analyze your backlink profile, identify potential landmines, and change your ratios to normalize your profile with that of your competitors.
Don’t follow in Elearners footsteps. By paying attention to the metrics analyzed below, you can know what behaviors to avoid to keep your site safe from the next dreaded Penguin update.
Step 1: Identify your Competitors (SEMrush)
Start by identifying the main competitors in the space
Quick Domain Compare (QDC)
How does your site compare to your competitors?
When the total backlinks, including links to subdomains, are analyzed, you can see that there are over 2 million backlinks, much higher than the other domains in the space. This is an instant red flag.
Quick Competitive Overview (CLA)
How does your site compare to your competitors in terms of Power and Trust?
Elearners has the highest LRT Power*Trust, yet it doesn’t have the highest number of root domains, indicating there are too many links from domains with high power or high trust.
Link Stats Comparison (Juice Tool)
Elearners is ranked #4 in terms of the number of unique C class links. Power and Trust is similar for all of their competitors (minus Walden.edu, which wasn't analyzed further).
Elearners has a normal distribution of Power and Trust.
Elearners also has over 14k keywords ranking in the top 20 according to SEM rush, making it #3 in this list. This should be an indicator of trust, yet you can see the steep decline in traffic.
You see in the chart below that Elearners has a very high ratio of sitewide links. This is another definite red flag.
How do various link factors compare?
In terms of Age, ACrank, PR, and Indexed pages, Elearners has a strong profile, similar to it’s competitors. Although the TitleRank isn’t that low, the fact that it isn’t #1 is a definite sign of a Google Penalty.
In all of these stats, Elearners is comparable, in fact it’s even stronger than most of the other sites.
Could a lack of social signals have been a factor in Elearners penalty? Even though Elearners has fewer than average Facebook likes, shares and comments, this isn't enough to prove significant. Strong social signals didn't prevent the site from contracting a penalty.
Summary of Findings from Quick Comparisons
At a quick glance, looking at these various factors yielded no significant findings. Elearners might have been a little off-balance in a couple of metrics, but there was nothing immediately visible to give us a concrete indication of why it suffered a Penguin 2.0 penalty.
Step 3: Detailed Competitive Analysis
In the next detailed step we’ll use the Competitive Landscape Analyzer (CLA) a lot more to get a better understanding of the playing field.
Link Status (CLA)
Are most links followed, nofollowed, or redirected?
Elearners has the highest percentage of follow links, which is often—especially from Google’s point of view—evidence of contrived links.
How are the links coded?
Looking at the Link Type metrics you can quickly see that Elearners has a major percentage of links from iframes. Why are there so many links in frames? More importantly, why is this number so high compared to their competitors?
This is something that definitely needs to be investigated as part of this link audit.
Deep Links Ratio (CLA)
How many of the links point to home vs internal pages?
Elearners has a higher deep links ratio than its competitors. Even though 5% is hardly significant, it stands out enough to call to question why this site is above average as compared to others in the niche.
Sitewide Links Ratio (CLA)
What is the sitewide links ratio of the inbound links?
Overall Elearners has a similar sitewide links ratio profile as other competitors, with the exception of a slightly elevated number of linking sites with 1-10 inbound links. This doesn't give us any conclusive information, however.
Referring Class C (CLA)
What is the distribution of the link popularity of the inbound links?
Here we see that Elearners has an unnatural ratio of links with more than 100K inbound links. While the average is 4%, Elearners has double that with an average of 8% of their links on sites with over 100K links.
Moz Domain Authority (CLA)
What is the distribution of the Domain Authority of the backlinks?
Elearners has a similar profile as other sites in the niche for Domain Authority.
Google Page Rank (CLA)
What is the PageRank distribution of the backlinks?
Elearners has an average PR distribution. They have 520 N/A links, one of the lowest of the group, as well as only 70 PR0 backlinks. On the high PR spectrum, they have 2 PR8 links, and 1 PR7 link, which is on the higher end of the average.
Link Velocity Trends (CLA)
How quickly are the sites building backlinks?
Elearners’ backlinks have a similar LVT as other competitors in the space.
By Retweets (CLA)
How active are the sites on Twitter?
Elearners seems to have a similar social profile to other competitors. No unnatural activity is apparent.
By Google +1's (CLA)
How active are the sites on Google Plus?
Again, Elearners seems to have a similar social profile to other competitors, and no unnatural activity is apparent.
TitleRank Home Page (CLA)
How are backlink sites ranking for their home page title?
Elearners has the lowest number of backlinks ranking #1. At 669, they are only at 52%, compared to the total average of 63%. Elearners also has the highest number of sites that are not ranking in the top 30 results (31% where the average is 19%). This is another red flag.
Link profile By PR & AC Rank (BLP)
What is the PR and AC rank of the inbound links?
Elearners has too many inbound links from sites that are not indexed in Google or have a PR or AC rank of 0. This is disproportionate to other backlinks as well as to other competitors.
Summary of Detailed Competitive Analysis
Did Detailed Comparisons yield any Red Flags?
We found significant findings in the following areas:
- Too many links in iframes
- Deep links ratio is higher than competitors
- Many of their inbound links have more than 100K inbound links
- High number of high PR links. Both of these indicate high Power in their backlink profile as compared to other domains.
- They have the lowest number of sites ranking #1 for their home page title
- Social Signals don't give us any conclusive information
- They have a very high number of inbound links from sites that are not indexed in Google, but the number isn't significant when compared to competitors
Step 4: Anchor Text Analysis
For a detailed, accurate anchor text analysis we use the Competitive Landscape Analyzer (CLA) a lot again.
Keyword Breakdown Money vs. Brand links (CLA)
What is the breakdown between Money terms vs. Brand and vs. Others?
To begin, we have to categorize the keywords into Brand, Compound, Money, and Other. This step can be time consuming but it is essential to the process.
Please note that you can use filters to bulk-check many brand or money keyword phrases by clicking the small “all” on top. Also notice that the first couple of keywords can already add to a 70% classification if they were overdone a lot. Make sure you classify 90% or more for accuracy.
The classification data is local to your user account and project, but Christoph mentioned that they are going to provide some great automation features in that area very soon.
Money Keywords (CLA)
What is the percentage of Money Terms in the Anchor Text Profile?
Elearners.com has the highest % for money, and lowest for Brand. This is a major red flag, for reasons we can identify when we look at the anchor text distribution of Elearners as well as some of its biggest competitors.
Anchor Text Distribution compared (QBL)
By using the Quick Backlinks Tool (QBL) or the more accurate Backlink Profiler (BLP) we can quickly understand the anchor text distribution by looking at the keyword clouds, weighted by number of links or by LRT Power*Trust.
What is the Anchor Text Distribution for Elearners?
Looking at Elearners.com anchor text distribution we can quickly see that they have too many money keywords in anchor text—the top 4, 5, and 6 keywords are money terms. This is an instant red flag that this is a contrived link profile with active anchor text manipulation. None of the densities are too high, but the overall density for "Money" terms is too high.
What is the Anchor Text Distribution for Devry (QBL)
In comparison we have Devry.edu. Notice their word map and how varied it is, focusing mostly on brand terms. None of the money terms show up in the top of the list for anchor density. This appears to be a very natural profile.
What is the Anchor Text Distribution for Kaplan (QBL)
Kaplan, on the other hand, also has money terms at the top of their anchor text profile. In fact, Kaplan is probably worse because the actual densities are higher. If this was thye only major issue for Elearners, then Kaplan would've gone down too. However, Kaplan is stronger than ever after Penguin 2.0.
Why didn't Kaplan get hit by Penguin 2.0?
What is the Money Keyword Distribution for Kaplan (BLP)
Since we already classified our keywords, we have them in all the other tools available for this project.
Why is Kaplan not penalized by Penguin?
I started by categorizing Kaplan's backlinks and performing a detailed link analysis.
Although Kaplan has money terms in the anchor distribution and the anchor density is high, the distribution between brand and money terms is greatly normalized—unlike Elearners.com. 64% of Kaplan’s backlinks are Brand links, as opposed to 34% for Elearners. By having a greater variety and variations of Brand terms in their backlink profile, they are protected from algorithmic penalties. A quick analysis into their backlinks also shows a great number of natural, unpaid links.
What is the Anchor Text Distribution for Phoenix (BLP)
Phoenix has the most natural looking profile, with lots of brand, click here, and organic terms. It's obvious that there's been little done to contrive this backlink profile.
What is the Anchor Text Distribution for Capella (QBL)
As with Phoenix, Capella has a natural and diverse backlink profile.
Summary of Findings of Anchor Text Analyis
What did Anchor Text Data Reveal?
- Too many Money terms in the Anchor Text profile
- Competitors that have high anchor text density were not penalized, possibly because of high Brand term density
- Anchor text word cloud looks very contrived for Elearners, with the smallest percentage of Brand Terms
Step 5: Link Detox & Detailed Link Analysis
Link Detox Overview (LD)
What is the Average Link Detox Risk?
According to the system, Elearners has a very low average link risk. This means that the bad links have been very well disguised in order to avoid detection. Keep in mind the average link risk here is still the average over all links.
Do any of the links stand out?
As soon as we check out the breakdown between healthy, suspicious and toxic links we see that they only have a mere 62% of healthy links.
Even though only 1% of the links are perceived to be “toxic,” we still have 36% of the links that are considered suspicious. These suspicious links may be where the problem is hidden. Now we'll take a look at some of these links individually for further information.
Scan Combined Backlinks (CLA)
Does anything jump out when you sort and scan through the backlinks of the group?
Download the CLA spreadsheet to Excel, and start scanning the backlinks.
I found a PR 8 to Elearners from StudyAbroad, and noticed that it's a"Partners link" at the footer. This is a footer link that's sitewide and available on every page of those 4 sites.
This is an indication of a potential network, leading to negative interlinkage and that co-citation with other link buyers certainly doesn’t help.
Looking at other competitor backlinks, many look natural. However, Elearners has many educational sites with keywords in the URL, which look unnatural.
Network alert! Network alert!
When further analyzing these links in the BLP link report, we can instantly see that many of these domains are owned by the same person, creating a link network. This is a HUGE red flag.
Identical sites on different domains
We also noticed that many of the sites are almost exactly the same, with identical templates and content, but with different domains and color schemes. There are other sites that are not quite as obvious but are still part of the same network.
Spot Test Elearners Backlinks (BLP)
If you spot test the links, do they seem clean/natural, or are they acquired/contrived?
First I sorted by PR, deleted all of the N/A's (links without any page rank at all, of which there are a lot!!! Another suspicious sign), and started spot-testing the high quality links. Here are a few examples of my findings:
Paid contextual links
All of the links end up at Elearners, which is obviously another site that is part of this link network.
Paid Links on USAToday?!
Even a link on USAtoday, which might've looked at first sight editorial, is purchased! You can see at the top of the page, the link to elearners.com
Summary of Paid Link Spot Tests
Why did the Link Profile look healthy?
Elearners hide their toxic links very well behind high profile / high quality paid links and link networks.
How can Link Detox identify very healthy, high quality sites as toxic links algorithmically?
This is, and has been, Google’s biggest conundrum when it comes to algorithmically fighting spam manipulation. Healthy links that affect pagerank and rankings are hard to identify without manual intervention.
So the question is, what can they do algorithmically to identify manipulated links? Look for unnatural ratios!
These unnatural ratios can trigger red flags and, when enough of the red flags are triggered, then an algorithmic penalty or a manual review can follow.
Christoph also mentioned some more improvements in this area, and similar to Google they are working hard on automating everything, but even Google needs manual review today still.
So what gave Elearners away, and caused the Penguin 2.0 penalty?
Take a look at this summary of my findings:
|Class C Popularity||ok|
|Backlinks To Domain||Over 2 Million Links||x|
|Anchor density||too many money terms||x|
|Link Health||36% Suspicious links||x|
|Link Type||Too many iFrame links||x|
|Deep Links Ratio||deep links ratio high||x|
|Sitewide Ratio TopDom||too many backlinks with more than 100K sitewides||x|
|Moz Dom Auth||ok|
|Google Author ID||ok|
|AC Rank Distribution||too many high acrank links||x|
|SEMrush Keywords||Unnatural distribution||x|
|Google Plus 1's||ok|
The X's are the number of strikes. Could it be that after a certain amount of strikes a site automatically incurs a penalty? Or could it trigger a manual evaluation, resulting in a slap?
So why did Elearners.com get a Penguin 2.0 penalty?
Too Many Unnatural Ratios
After analyzing about 20 factors, we found red flags in about 10 of the different items.
Too many links with Money Terms in their Anchor Text
Their anchor text profile shows a large number of money terms, higher than other competitors in the space. Simultaneously, the number of Brand terms is lower than other competitors in the niche. Looking at their anchor density word map also shows that there are few "noise" keywords, thus showing a contrived backlink profile.
Too Many Paid Links
Given that many of their high PR links are paid links, these may have been identified by the algorithm or a manual review, resulting in the penalty. By penalizing Text Link Ads and their network, Google is making it clear that they have no tolerance for people buying or selling links. Spot testing their backlinks shows many paid links, with just a few examples below:
Elearners is part of a Link Network
Many of their inbound links are part of the same network, many registered by the same person, others hidden behind different registrars, even more hidden behind private registrars. Upon inspection it's fairly obvious that they're owned by the same company.
It's likely the network started years ago with them buying high PR links, which earned them visibility. This visibility led to some natural links, including links from some .gov and .edu sites. From this authority and pagerank, they continued to create more sites to create a large link network of sites, all interlinked or randomly linked.
This network includes hundreds of niche sites, each focusing on specific degrees. By linking within the network using footer links or iframes, all of those sites gained high pagerank.
By looking at the BLP backlinks and investigating each of these network links, many of them retain pagerank, titlerank, and SEMrush keywords, therefore the entire network hasn't yet been popped. Many of the sites continue to thrive and feed the main site, Elearners.com
Protect your site Against Penguin!
So what does this Penalty tell us about Penguin 2.0?
Watch your Ratios!
As evidenced by this study, it is vital to keep an eye on all of your ratios. If too many of your ratios look unnatural as compared to others in your niche, these red flags may result in a manual review or automatic penalty.
Watch the number of Money Terms in your Anchor Text
It's not enough to just watch your anchor density—you also have to watch the percentage of money terms in your anchor text. Study other competitors that have healthy, natural link profiles and emulate them. Or, better yet, follow their same tactics to acquire natural links with natural anchor text.
Use Brand and Noise Terms in your Anchor Text
Try working on link building without contriving your anchor text. Allow people to link to you however they want, to result in natural looking links.
Don't Buy Links!
Buying links worked for years, but Google knows this is a weakness in their algorithm. By using Penguin with a combination of manual reviews, they are now able to penalize sites that are buying links.
You may buy links and get away with it for a time, but eventually it's possible that your link buying may trigger a penalty, causing your site to tank in the rankings. And, as many people know by now, once you have a Penguin penalty it's very difficult, almost impossible, to recover.
Avoid Link Networks
It's very tempting to buy into a link network, or to create your own network of niche sites. Many people do it by buying expired domains, or by finding established networks and joining. This may work for a time, but eventually some of these ratios will be triggered, and the network will be found. Once you catch the tail of a network, exposing the rest is fairly easy.
Network builders try hard, but there are always footprints left to find, and with the sophistication of Google's algorithms, you better believe the network will be identified and penalized.
Don’t Procrastinate! Do your Link Research TODAY!
To algorithmically monitor for spam, Google looks at your site as compared to your competitors. If your site sticks out with many metrics outside of the norm, it may be a call out towards a penalty. One trigger is not enough—as we saw with the comparison to Kaplan and its high density of anchor texts. One signal didn’t lead to a penalty; having many unnatural ratios can. Ratios are increasingly important as Google looks deeper into unnatural link building and controlling spam.
What does this mean for your site? If you’ve already been penalized, run an audit with LinkResearchTools and look at your link ratios to see what you can normalize. Watch your rankings and traffic from Google to see if it normalizing helps your site recover and perform better.
If you haven’t yet been penalized, protect your site by continuously running these Link Research reports to keep your ratios safe. Be extra-vigilant in your optimization efforts to make sure that you are not triggering red flags. If you are prepared and avoiding triggering linking behaviors, you won’t have to worry when the next Google Penguin update comes around.
This case study was written by Marcela de Vivo, CLRTP and was reviewed and approved by Christoph C. Cemper for publishing as Certification work for the Certified LRT Professional level.
A word from Christoph C. Cemper
This analysis was conducted and post written by our LRT Associate Marcela De Vivo, CLRTP.
Marcela really went to great lengths with this case study and I'm happy to certify her for our Certfied LRT Professional level by approving and publishing her research on our site.
This is Marcelas next step towards the Certified LRT Xpert level which is pre-requisite for the Certified LRT Agency certification. Both will qualify her to receive consulting leads from us. Our goal is to provide our community and clients a high quality service, and our certified experts are key to that.
I am sure Marcela will be there soon. As you can probably judge from her work already, I can already wholeheartedly recommend her to work with you whenever you get a chance!