When it comes to website design and messaging, nearly everything you read encourages testing. You don’t generally see the same enthusiasm for experimentation in SEO. Why is that?
If you believe all that you read, optimizing for search can lead you down a path of appeasing Google’s guidelines, rather than creatively finding ways to increase visibility in SERPs.
SEOs too often take what Google’s Webspam team and authoritative blogs broadcast at face value and optimize to “best practices” accordingly. I’ll admit that I’ve been guilty of drinking the Kool-Aid plenty of times without doing any critical examination.
I think Bill Sebald of Greenlane Search Marketing sums up my point best:
The truth is, without testing you can’t know definitively what works and what doesn’t. In a world where many SEO strategies have shifted focus to pure content creation, it may be your unique insight from consistent testing and tweaking that gives you the edge in SERPs.
The remainder of this post will give you ideas to test while citing interesting outcomes of search experiments and case studies to get you started down the road of SEO experimentation.
Set Up a Testing Environment
Unless organic traffic and search rankings aren’t important to your web presence, it’s best to set up one or several websites on which to test your SEO hypotheses.
If you own any now-dormant domains, they may be perfect candidates for revival as a test website. Otherwise, you can quickly and cheaply register a new one and spin up a WordPress site.
The tests themselves can be created and measured with this simple procedure:
- Create a control page that targets a specific search query. It’s a good idea to target a low volume, low competition search term or you can make up your own word, like “zizzantyx”.
- Spin up nearly-identical duplicate pages that vary by a single factor.
- Monitor the crawling and indexing of these pages over 10 or more days and watch for which ones outrank others.
Test #1: What Does a Perfectly Optimized Page Really Look Like?
The visibility of your great content is likely to be affected by its level of on-page optimization. But which elements have the greatest impact and how should they be implemented on a web page?
Your on-page SEO checklist likely addresses the following:
- Title tag and meta description
- Heading tags
- Body text
- Internal links and external links
- Page URL
- Image title and ALT text
- Additional markup
You should test not only the usage of specific on-page elements, but also their location in the context of the page and positioning of the keyword within the elements themselves.
Testing “Old School” SEO Tactics
Bill Sebald of Greenlane Search Marketing tested several “old school” SEO tactics against one of his monitored keywords, “philadelphia seo company”. His tests improved his rank by three positions and he found that:
- Exact match anchor text from a single page that passes PageRank still has an impact
- A sitewide link with exact match anchor text actually hurt his rank
Bill also tested his title tags to see if length or keyword position would make a positive impact. It didn’t. Best to stick with relevant title tags with the goal of providing a good user experience, he concluded.
The Best Performing Anchor Link is One You Aren’t Using
Dan Petrovic, known experimenter with Dejan SEO, ran a similar anchor link test comparing four different styles of exact match anchor text. Interestingly, the winner was a page linked to by an image with ALT text containing the target keyword.
Additional On-Page SEO Testing Ideas
- Which elements are best to optimize for location-specific keywords to to make the biggest impact on local search results?
- Does schema markup of a business’ name, address and phone number impact a site’s ranking ability?
Test #2: What is the SEO Value of Low-Quality Backlinks?
Backlinks are an SEO’s most loved and feared ranking factor. While the strong signal of a quality backlink can greatly improve search visibility, the murky classification of poor backlinks and their potential downside lead many to stay away from link building altogether.
However, if you’ve ever done a competitive audit of high-ranking sites in SERPs, you’ve definitely run across some suspicious backlink profiles. This can be especially frustrating for those of us that are focusing on white hat SEO tactics, yet are finding lower quality sites appearing above us.
Google’s definition of unnatural links covers a variety of linking scenarios. Some of the most frequently cited are:
- Quid pro quo link exchanges
- Buying or selling links
- Automated link building
- Low-quality directory or bookmark sites
- Widely distributed links in headers, footers or widgets
- Links in forum comments
Although having these types of backlinks can expose a website to algorithmic or manual penalty, that doesn’t necessarily mean they can’t positively impact search rankings.
Nofollow Backlinks Do Have SEO Value
Conventional SEO wisdom says that nofollow links do not impact your standing in search results. However, Rand Fishkin of Moz found that they did.
Rand ran multiple tests in which they built in-content nofollow links from more than 40 domains. After the links were indexed, the target web page’s ranking improved by one to three positions.
Need further proof? Even though nofollow links do not pass PageRank, according to Google, they have said that they will take action against sites that pursue mass nofollow linking. I would speculate that this is, at least in part, due to the ability of nofollow links to impact search results.
Rand’s backlink experimentation had an interesting and unexpected side effect. Months after removing the backlinks, their rankings continued to climb. Some pages went as high as #1! This was attributed to a phenomenon called link echos, whereby disappearing links don’t correspond with a drop in rankings.
Black Hat Backlink Strategies Are Still Working (Temporarily)
It’s the most widely-known secret of the search industry. SEOs are constantly finding and sharing examples of black hat link building leading to dominance in search results. Despite Google’s stern warnings, it appears that their capacity to detect and clean up affected SERPs isn’t quite at the level that they lead us to believe.
Matthew Barby, a digital strategist at Wyatt International, uncovered a startling example of black hat SEO at work. The offending site was able to outrank industry leaders for the term “seo new york” in just three months using spammy backlinks. Although the publicity doomed them, they were likely generating thousands of unique visitors and plenty of leads during their reign.
More recently I conducted a backlink analysis on behalf of a local client seeking to rank for a Minneapolis-related keyword. On the one hand, there were a few well-established businesses with quality backlinks acquired through legitimate content marketing and PR. Above them ranked a newer competitor whose backlink profile was almost entirely comprised of low-quality directory sites with exact match anchor text.
Noted local SEO expert Mike Ramsey has lamented that with the introduction of Pigeon, “spam works better than ever” when it comes to inflating local search results.
The black hat approach to link building, in most cases, produces short term results. Rankings are boosted for several months before dropping. Although the gains are short-lived, the temporarily lift is enough to produce ROI for some. Unfortunately. this means that spammy link building will continue to influence SERPs.
Additional Backlink Testing Ideas
- How many low-quality backlinks does it take to compete with one high quality backlink (yep, I’m testing that)?
- Do exact match anchor links from press release sites still provide SEO value?
Test #3: Get Yourself a Manual Penalty
A manual penalty from the all-powerful Google seems like the most ominous threat to an SEO’s livelihood. It can obliterate your rankings and drop your organic traffic to zero. The reality? It’s not that bad.
In most cases, they don’t cause a total loss of rankings and traffic, but rather a demotion. While there are plenty of horror stories of Google’s Webspam team not replying to reconsideration requests, it usually takes anywhere from a couple days to a couple months for the penalty to be removed.
If you’ve never experienced a manual penalty for yourself, use your testing environment to give it a try. You’ll learn the cleanup and review process firsthand and be able use that knowledge to address a real penalty if you ever incur one.
I reached out to Cyrus Shepard, who has written extensively on penalization and recovery, for his advice on how one can quickly ensure a manual penalty.
I posed the same question to Dan Petrovic, mentioned earlier in this post.
You may find that a manual penalty is harder to come by than you think. But if you follow the suggestions above, you’ll eventually get this notice in Google Webmaster Tools.
You Should Be Concerned About Negative SEO
Yonatan Dotan, head of SEO at yellowHEAD, recently had to help a client through a scary manual penalty. Overnight, they lost 94% of their organic traffic, including traffic from branded terms.
The client got caught in the crossfire of spammers trying to build links to their own sites. They sneakily included links to legitimate sites in an attempt to fool Google into thinking that the backlinks were originating from a quality source.
Yonatan points out that, although his client’s penalty was coincidental, if their client had been specifically targeted in this link scheme, the offenders would have successfully knocked them out of search results.
Google, up to this point, has acknowledged search professionals’ concerns over negative SEO, but will only say that algorithm updates continue to lower the risk without going into specifics.
Thankfully, the yellowHEAD team was able to recover 82% of their client’s organic traffic after waiting a month to hear from Google. However, the question remains: If you can intentionally earn yourself a manual penalty, why couldn’t you repeat the same process for a website that you don’t own? Perhaps your testing will give us all some insight.
Additional Manual Penalty Testing Ideas
- Which tactic on it’s own is most likely to get noticed and cause a manual penalty? Well known link-networks, hidden text, links from penalized sites, or blog comment spamming?
- How long does it take for disavowed backlinks to disappear from your backlink profile?
Go forth and test SEO conventions. Look at all claims with a degree of skepticism and confirm or deny them through your own testing process.
Most importantly, share what you find. Results from your experiments may help the SEO community uncover or support new trends, algorithm changes, and more.
Note: The opinions expressed in this article are the views of the author, and not necessarily the views of Caphyon, its staff, or its partners.