The last quarter of the year is here. It’s time to ask yourself if you’re beating your personal records from last year or losing ground. If it’s not looking good, this roundup may be the boost you need to clear the final stretch and come out a winner.
In this month’s guides, we’ll cover Ryan Stewart’s 1-day SEO hack, a breakdown of the three stages of link building, a new method for finding local SEO keywords in bulk, and a neutral look at whether PBNs are still performing.
Then, you can catch up on the latest research with case studies. You’ll see the data on building backlinks with a stat page, a breakdown of Google user behavior, and whether your SERP position affects how much room Google gives your description.
We’ll close with the news. This month, a college kid fooled tens of thousands of people with an AI-generated blog, and Apple made moves that suggest it may be launching a search engine.
Are there any hacks left in SEO? Ryan Stewart often says ‘no,’ but now he thinks he found one. He’s bringing us a technique that he used to create 1-day growth on three different test websites.
Ryan theorized that if he purchased an old post on one website, then redirected it to a new copy of that post on a different website, the new page would inherit the old one’s power.
He based this technique on an earlier test by Brian Dean (who authored one of the upcoming case studies this month). Brian bought an entire SEO website to merge into his site, but Ryan wanted to prove that even a single page could work.
Ryan started with the top-performing post for a keyword he desired. The process was simple. He pitched the current owner, asked if he could buy the whole post for his site (priced at $650), and requested that the original page be 301’d to the article’s new home.
When the original owner agreed, the new page on Brian’s website experienced this in one day:
That may sound easy, but he cautions us that it’s not. For one, it’s expensive. Anyone giving up a post strong enough to pass this kind of power will expect a fat check.
Additionally, this sort of redirect operates like an endorsement. There may not be a price a website owner will accept if it means sending their nurtured traffic to a sketchy site.
Don’t worry. As he explains, you can still work this strategy by doing something as simple as buying a site at auction.
It’s a great hack, though it’s expensive in proportion with its value. If you prefer cheaper DIY strategies, the next guide will help you do that with local keyword research.
If you don’t have the budget to acquire these types of links, don’t worry. Our next guide will introduce you to a lot of new options as you learn what to build along the 3 stages of link building.
3 Stages Of Link Building: What Links To Build And When
https://authority.builders/blog/3-stages-of-link-building/
Authority Builders brings us this deep look into the why, how and when, of building links in 2020.
It starts with a theory on how we paint the picture that we want Google to see when it looks at our links. That theory goes like this: Each backlink is graded by the power, relevance and trust it signals to Google.
Power is measured by how authoritative the linking page is on its own. This is difficult to measure now that PageRank is no longer public, but we often still see the effects of it working when we build links.
Relevance is measured by how contextually appropriate the link is to the page. Relevance comes from:
- The niche of the linking page
- The overall niche of the linking domain
- The niche of the anchor text
Trust is the most difficult to measure, but this article has two possible theories for how it is measured.
The first hypothesis is that Google relies on (possibly handpicked) “seed sites” that are older, heavily trafficked, and known for high-quality content. These sites are marked as trusted, and the sites they link to inherit some of that trust, which they can then pass on to other sites.
The second hypothesis is that trust is a byproduct of ranking. Domains and pages that can be ranked (fewer than 10% of all existing sites) pass on trust to the sites that they link out to.
Using these three priorities, Authority Builders then breaks down three stages of a site’s life, and what matters most. The stages are marked by:
- The Sandbox: This stage covers when your site is brand new, and truly authoritative links look suspicious. The theory goes that the best links to build in this stage are relevant guest posts because they’re brand new content (like your site).
- Trustworthy stage: This stage covers the time when your website has established itself as helpful to its audience. In this stage, you should focus on power, by doing outreach to sites that are already powerful. Link insertions are a good way to get into better sites.
- Authority mode: This stage covers your site as soon as new content placed on it can start ranking immediately. This is the time when trusted top-tier editorial links start to matter. They can be built by investing in shareable content and media mentions.
This blueprint is measured with a case study in the article that shows a brand new site brought to the point that it can attract 120 quality backlinks per month.
The results are great, but following this blueprint will take a couple months. The next guide will help you make some progress in days with local keyword research.
How to Identify Local SEO Keywords in Bulk
https://localu.org/how-to-identify-local-seo-keywords-in-bulk/
Deciding which keywords to target for local searches can be a hassle, especially if you’re trying to help a company with dozens of locations and a variety of products/services.
Digging for the keywords that return local results can take hours, but Fion McCormack may have a solution for us.
His method starts with an unstructured list of all the keywords you want to test for local relevance. You can grab these from whatever tools you use (SEMrush, Ahrefs, etc.) to find bulk keywords.
When you have them, you’re going to:
- Convert that list of keywords into Google search query URLs (create the URL that is produced when you run the search through Google)
- Import the URLs into Screaming Frog
- Set a unique identifier (Fion used the “more locations” element that Google places at the bottom of all map packs)
- Run SF on your URLs to automatically filter the list by local results only
That sounds easy, right? It is, but only because Fion was nice enough to provide us with all the sheet templates, Excel functions, and SF commands that turn this job into a breezy string of copy-pastes.
It’s always great to learn a new technique that makes your work simpler without getting a speck of dust on your white hat. In the next guide, though, we’re going to wade into an issue that’s a point of contention in the SEO community—The world of PBNs.
PBNs: Everything You’ve Ever Wanted to Know But Were Afraid to Ask
https://www.searchenginejournal.com/private-blog-networks/377296/
If you read this blog regularly, you know that I’m no stranger to PBNs. I’ve done large-scale PBN case studies and covered techniques you can use to measure the link juice PBNs provide and test PBNs before adding them to your network.
I’m not that neutral on them, so my readers may find Jeremy Knauff’s coverage in this guide refreshingly even-handed. He covers the questions most newbies are nervous to ask, along with the technical details and pervasive myths that surround them.
Pre-penguin, there was not much debate at all: PBNs were some of the best links you could build for the money. Where they are now is the focus of this guide.
First, he establishes that PBNs are still very much with us. They have changed a bit. Today, they tend to be “living” sites with high-quality and regularly-upgraded content.
As a result of this change, Jeremy points out, many people are using PBNs without being aware of it. Some of the most authoritative guest posting services use them, and many niche sites you might be tempted to do outreach to may also secretly be PBNs.
From that understanding, the guide confronts some myths:
- Myth 1: PBNs don’t work: They do, and they may already be working for you.
- Myth 2: They’re garbage: They’re nearly impossible to tell from real sites when they’re built correctly, even to a human visiting them.
- Myth 3: Google can easily identify them: Avoiding the same registration info, hosting, themes, content, and links will make them look and act like a normal site
In the end, Jeremy refrains from saying whether or not they should be used other than to say that you should know what you’re doing if you try to begin. That goes for many techniques.
That’s the last of the guides, but this month’s case studies are nearly as actionable. Let’s start looking at an experiment where a “stats” page was set up and claimed a #1 spot in SERPs.
Link Building Case Study: How We Built Backlinks With a ‘Stats’ Page (And Ranked #1)
https://ahrefs.com/blog/link-building-case-study/
Anyone who writes about SEO frequently will find themselves searching for “SEO stats” to reinforce a point or defend something to a client. The SERPs page for that term looks a little different today because Ahrefs decided to claim the #1 spot for it on a whim.
It started with an SEO statistics post they put together back in June. This normal-looking post was, in fact, a test case for a novel type of outreach campaign.
The test was based on the Skyscraper Technique. This is a process where you search for a page with lots of links, build a better one, and then pitch the new and improved version to the people linking to the original.
It’s a simple value proposition. Everyone wants their visitors to consider them credible, and linking to a better resource instead of a worse one is a simple choice.
Joshua Hardwick of Ahrefs decided that this choice was even more straightforward with statistics pages because they tend to have a ton of traffic and links despite being—on average—sloppy and outdated.
Joshua and the Ahrefs team decided that these conditions could be quickly leveraged into a top-performing post by:
- Finding a winning angle for the stats page
- Tracking and refining the link prospects
- Creating the new and improved stats page
- Sending out outreach emails
The test post was only published at the end of July. It’s now #1 in SERPs after attracting 27 links that used to go to lower-quality posts. That number only includes the links that outreach generated.
Before the test even ended, uncontacted websites were already building links to it.
By the end of the test, nearly a dozen DR70+ domains had built links to the stats page, along with another dozen DR40-69 domains.
This was an impressive confirmation of the skyscraper technique in action. If you want to develop techniques like this yourself, it helps to understand user behavior. Just in time, Backlinko has some fresh research for us.
How People Use Google Search (New User Behavior Study)
https://backlinko.com/google-user-behavior
If you’ve been studying how people use search engines since the early days of the internet, you’ve probably noticed it can change fast.
People can teach their own eyes to filter out things that don’t provide them with value, and many old types of advertising have died out as a result.
This research by Brian Dean could help us catch up to where users are now. Among other conclusions, the research showed that:
- 9% of searchers make it to the bottom of the first page of the search results
- 7% of users bounced back to the search results after clicking on a result
- Only .44% (note the decimal) of searchers go to the second page of Google’s search results
- Only 3% of Google Users Interact With a People Also Ask Box (nearly half of all searches display this box)
- Searchers use one of Google’s autocompleting suggestions 23% of the time
The study was pulled from actual human behavior rather than a survey. The keystrokes and actions of 500 people were tracked as they looked for commercial products, services in their area, and products for their needs.
It’s worth searching the whole list for data that may be useful to your online business.
We’re going to scale down a little for the last case study. This narrow experiment simply looked at how much room Google gives descriptions, and whether that amount changed depending on rank.
Does Position impact meta description length on SERP?
https://www.rankranger.com/blog/meta-description-length-by-position
Meta descriptions have a weird place in SEO. They make an important impression on searchers, but otherwise, they aren’t a direct ranking factor. Also, we can’t stop Google from simply replacing them at will.
If it’s mostly out of our control, is it possible to use them to claim an advantage? Maybe it is because Google may allow higher-ranking pages more room to make a case to searchers.
Mordy Oberstein of Rank Ranger put together these results after tracking 5000 keywords for a month. That’s a ton of data points for a question like this, so the results should tell us a lot.
What he discovered was that there was a difference. The first five results (on average) got to enjoy 15 more characters than mid-page results, and up to 18 more characters than the lower results.
As Mordy points out, this raises some interesting questions about why CTR decreases as searchers move down the page. It isn’t just position—it’s that they’re given less room to explain themselves.
It seems we’ve learned another reason to fight for the highest position possible.
That closes out our case studies for the month. We’ve got some fun news for you, including a story you may have missed about how a college kid’s fake AI-populated blog became the day’s biggest story on Hacker News.
A college kid’s fake, AI-generated blog fooled tens of thousands. This is how he made it.
https://www.technologyreview.com/2020/08/14/1006780/ai-gpt-3-fake-blog-reached-top-of-hacker-news/
Just when you think the wild-west days of SEO are over, here come some young up-and-comers to remind us that there are always new tricks to learn.
As Karen Hao covers in this piece for the MIT Technology Review, a student named Liam Porr used an AI model to create a fake blog. It wasn’t supposed to go much further than that—but suddenly, one of the posts his AI churned out hit the #1 spot on Hacker News.
Of course, this wasn’t just any AI. The post was written by a powerful language-generating model known as GPT-3. The model is only available to some researchers and students at this time.
The AI was given a headline (“Feeling unproductive? Maybe you should stop overthinking) and an introduction in the self-help and productivity genre and produced the rest. For a lot of hacker news readers, that was enough.
As the student himself admitted, it wasn’t a great article. He chose the topic because it was too personal for readers to expect any verifiable facts or logical progression.
However, the ease with which an AI can produce otherwise-attractive spam is something that may be keeping the people at Google up at night.
The people at Google may also have some other reasons to be losing sleep. The behemoth brand Apple has signaled that it may be ready to break into the search engine game.
Apple showing signs it may soon launch a search engine to compete against Google Search
https://www.coywolf.news/seo/apple-search-engine/
Is Apple about to launch its own search engine? Right now, we don’t know much, but there’s some evidence to consider.
First, the exclusivity agreement between Google and Apple (estimated to be worth billions of dollars) that makes Google the default search engine on Apple devices is coming to an end.
Join the discussion here on Facebook.
