What Thousands of Web Marketers Get Wrong about SEO
Like many other people I do my best to share what I’ve learned about the Internet, Web marketing, and SEO in general with people via Web forums and social media. I don’t respond to most questions I see for a variety of reasons, including the all-humbling “I have no idea of how you do that/why that happens”. No one in our industry knows everything. No one even comes close to knowing everything.
If anyone with years of experience decides to help you with a problem, the best they can do is tell you what they believe worked for them. Many of us often share less than our best advice because we still remember what it was like to feel lost, uncertain, and like taking the next step was a huge gamble. That’s why we tend to include “it depends” in so many of our answers. That’s why we sometimes say, “what works for me is …”
Somewhere along the way people forget that “what works for me” isn’t the same as “this is how it works”. What works for me means “I did this and subsequently [something] happened.” On the other hand, this is how it works means “if you do X then Y is going to happen.”
There is nothing subtle in the difference there but to some people when you point out these differences they feel like you’re telling them white paint isn’t white any more. And their responses to gentle (or rough) correction tend to range from spewing conspiracy theories to launching vitriolic attacks against your worth as a human being.
Here are some of the worst ideas still being passed around on Web marketing forums and social media in 2020.
1. You Can’t Trust What Google Says But …
Last time I checked Google had not published a comprehensive guide to its search engines’ algorithms either via some Webmaster blog post or a research paper/patent you can find via Google Scholar.
Amazingly, you don’t have to look long or hard to find someone who angrily proclaims these incontrovertible truths:
- You can’t trust what [Googler name] says because they’re all lies
- Google’s algorithms do [X] (because Googlers say the exact opposite)
- If you don’t believe either point above, just read (some popular, well-known marketing blogger who openly challenges and contradicts what Googlers say)
You know who these bloggers are. Some of them are even quoted in the news media. Heck, Google even links to some of their articles as if they are telling their readers what Google tells people.
It’s no wonder the marketing public are confused about what to believe. Google clearly isn’t helping the situation. But I digress.
Many ex-Googlers (or Xooglers) have left the company and launched or joined digital marketing companies. They share tips on social media, sometimes write interesting blog posts, and even speak at conferences. Far too few of them are sought out or quoted by the news media, when the media publish articles about the woes of Web marketing.
The people who are most likely to be cited as “experts” in search by the media and a majority of random voices in online discussions have these qualities to validate their pronouncements on all things Search and Google:
- No formal training in computer science or information retrieval
- Minimal grasp of basic mathematics
- Surface knowledge of statistical analysis
- No inside knowledge of what the algorithms do
- An addictive-like dependence upon 3rd-party SEO tool data
In other words, the people most likely to say anything influential about how search works and what Google’s algorithms actually do are the people least knowledgeable and least qualified to teach anything about how search works and what Google’s algorithms actually do.
Not everyone is bamboozled by these charlatans but someone who recently trashed a couple of them in social media comments didn’t give credit where credit is due. These people become popular and influential because they work hard at becoming popular and influential. Their theories about how search works may be nonsense but they can still add mailing lists to their Websites, they still leverage their industry and media connections for links and publicity, and they push their content via social media like carnival hawkers screaming at passersby.
Their basic marketing methods work even though they have no real effect on search results. It’s that success more than anything else that feeds the false credibility of the SEO mythologies that create havoc among businesses just trying to get some traffic from Google.
2. Everything You Do Should be about E-A-T and YMYL
I actually see more of this bullshit in YouTube videos and blog posts than I do in Web marketing forums and social media. Maybe I just don’t visit the right marketing forums.
Google’s Quality Rater Guidelines are still widely misused by people teaching others how to optimize for Google. And you can see how ineffective their EAT- and YMYL-based theories are because of the frustration that their would-be disciples express when asking for help.
You know what people have been reading (or watching) by the buzzwords they use when asking questions. And some of these questions are small essays that manage to include at least half a dozen SEO metrics, references to the Quality Rater Guidelines (without mentioning the QRG document), and at least one or two SEO conspiracy theories concerning CTR, bounce rate, or “social signals”.
Somehow, magically, despite all of Google’s denials and repeated clarification from the supposed experts on E-A-T that these concepts aren’t in any way enforced by Google’s algorithms, people continue to ask how to optimize for or improve their E-A-T.
And everyone has a Your Money or Your Life Website. If you read the latest version of the search quality rater guidelines (link opens in new tab), you’ll see that the definition for YMYL sites has exploded to cover about half the Web, and is so generic as to be useless.
Just today I watched a recently uploaded video where the YouTuber described a fitness blog as a YMYL site. It’s an affiliate site so it’s not asking for any money, and it’s not the kind of site that someone suffering a horrible disease needs to read.
WebMD is a YMYL site. A fitness blog earning money from affiliate links is not that important. In any event, Google doesn’t have algorithms (that they have publicly mentioned) that label sites as “YMYL”.
There are far fewer YMYL sites out there than the Web marketing community has convinced itself must exist. But here’s a bit of advice for everyone who believes they are running a YMYL site: If you’re having THAT much trouble getting traffic, pick a less competitive niche.
If you’re just blogging for affiliate referrals, you can promote just about anything and make money. Those of you who (like Reflective Dynamics) see a lot of referral sales know what I mean. When you see 1 visitor drop $1000 on obscure tools and equipment you’ve never heard of in a single purchase event, suddenly you realize that YMYL is mostly bullshit.
When people are ready to spend money they spend it. All that matters to the affiliate marketer is whose cookie or site ID gets the referral credit. You guys are worried about finding good writers to turn out 300-word product reviews. You have no idea of how ecommerce actually works.
People come for the poems and buy antique car engine kits.
You don’t have to be an expert in anything to make money on the Internet. You need traffic and you need good, reliable merchant networks. The rest is up to the Web surfing public and you have no control over those people.
3. Everything Google Does Is about BERT (or some other algorithm)
If you read a blog post about some interesting Google patent, that doesn’t mean you need to drop everything and start manipulating whatever experimental signal the patent may be describing.
Most of the patent analysis articles I’ve read have totally missed the mark on what the patents were describing. While it helps (considerably) to have enough of a background in formal computer science and/or information retrieval to understand a lot of the processes these patents describe, the real secret to understanding them is to research who filed the patent.
I don’t mean the company. I mean the person whose name is on the patent.
At least half the articles I’ve read about Google patents over the past 10 years analyzed patents that were filed by Google’s AdWords teams. And yet those bloggers assumed everything in the patents pertained to Google’s organic Web search.
It doesn’t work that way.
Several times a year I find myself discussing how Google uses CTR and bounce rate with people who pull out their favorite blog posts as proof they know what they are talking about. Again, about half the time the patents described in the blog posts describe advertising network processes.
It’s not always clear to me what these patent filings describe, but I usually learn what I need to know by checking out who wrote the application. When it turns out to be someone from the advertising services I look a little deeper – just to be sure – to see if they ever worked on the organic side of search. A few advertising engineers have worked both channels. Some engineers develop processes with teams from other channels, too.
But most of these patents (and research papers) describe very specific things that are only relevant to one channel. And for some unlucky reason many SEO bloggers mistake advertising network patents for organic search patents.
Of course, for the past year or so BERT has been a popular topic of misguided analysis. I’ve never seen so many machine learning algorithm experts with no background in machine learning in my life.
As a rule of thumb: If you intend to explain something to the masses, you must understand it well enough to explain it to an idiot. That’s easier said than done. It’s a paraphrase of physicist Ernst Rutherford’s statement that “an alleged scientific discovery has no merit unless it can be explained to a barmaid”. You may have encountered the more hip version that says “if you can’t explain it simply you don’t understand it well enough”.
I don’t know much about barmaids in Rutherford’s day but I suspect far fewer of them were college-educated women, so his statement may not have been as sexist as it sounds. I’ve known some very well-educated female bar tenders and they can hold their own in scientific conversations.
Most Web marketers, on the other hand, can’t.
When Google announced “the BERT update” they said that BERT only influenced about 10% of all queries. Immediately the Web marketing world set about the dire task of explaining what BERT is and does to each other.
They got it wrong.
BERT doesn’t have anything to do with crawling, indexing, or ranking Websites. BERT teaches other algorithms what words and phrases to expect when they see certain other words or phrases. BERT is the teacher and some of the algorithms that process search queries are BERT’s students.
If you don’t understand BERT well enough to grok that, you have no idea of what BERT is or does.
4. These are the Signals Google Uses to Rank Websites …
You’d think after 12 or so years people would have grown tired of believing this nonsense.
The SEO community does not have a list of Google’s algorithmic signals – the so-called “ranking factors”. No one has this list.
SEO bloggers love to write up lists of “200 ranking factors”. I’ve seen some very well-documented lists, too. They link to Google blog posts, Google patents, Google research papers, Google Tweets, etc.
The idea that you must be right because you can link to some Google statement about a “signal” is absolutely ridiculous. Not one of these lists of signals explains how the signals are used, when they are used, or why the use of a signal may not mean anything.
Yes, a few writers toss out caveats and disclaimers. That’s not what I mean.
You should be able to explain how a search engine selects and scores results before you start telling the world which signals it’s using. Even if we could assume for the sake of discussion (and it would be stupid to do so) that you have an accurate list of signals, that list by itself doesn’t do anyone any good.
Sadly, every year some popular SEO blogger publishes yet another “correlation study” implying proof that their theories must be correct. It was once easy to explain how these correlation studies don’t work. But now the bullshit artists have returned with LARGE VOLUMES OF DATA.
Correlation analysis – applied to reverse engineering search engine algorithms – is still complete bullshit. It doesn’t tell you anything about what the search engines are doing with their data.
But apparently if you scale up from “I have traffic estimate charts here for 2 Websites and therefore conclude there was a MASSIVE GOOGLE UPDATE” to “we crawled 100 million Websites and found these correlations”, people are satisfied that you just did something scientifically valid and computationally meaningful.
If correlation analysis could really pull that rabbit out of the hat Google would have done something to break the correlations years ago.
5. Has Anyone Tried to Do [some brilliant idea] …
The thinking behind these questions is that “I’ve just discovered a really clever way to hide something from Google”.
The better questions ask about things that are not explicitly mentioned in the search engines’ Webmaster Guidelines. If there is no specific guideline against it, then it must be a good idea, right?
When Larry Page and Sergey Brin wrote their Backrub paper in 1998 they got A LOT OF THINGS WRONG. Time and the Internet have proven just how much bullshit went in to building up the mythology of PageRank.
Even so, Google has been around for more than 20 years, it employs tens of thousands of highly educated, really smart people, and they’ve completely rewritten the organic search system several times.
I have seen maybe 1 brilliant idea for the first time in the past year. People need to accept that Google has seen all these clever strategies before.
In many cases they don’t have to do anything at all because people don’t realize just how equalizing a massive search index can be. And yet Google probably incorporates hundreds of document classifiers in its algorithms today just to ferret out all the clever ideas that actually worked for a while in the past.
The real lesson to be learned here is not that you can sometimes fool Google (you CAN) but that it probably won’t last forever, especially if you blab about your success on blogs, at conferences, on social media, in Web marketing classes and tutorials you sell to the public, and in YouTube videos.
The more you talk about it, the sooner it dies a painful, expensive death.
If you figure out a way to get past Google’s defenses and want to know how to monetize, rule number 1 is “Keep Your Mouth Shut” and rule number 2 is “Don’t Share It With Anyone”. Rule number 3 is “I told you so, you stupid idiot.”
People invest so much energy in trying to find effective short cuts for Google SEO they never stop to realize how much time they waste when all they have to do is create the kinds of Websites they themselves would be happy to recommend to other people in search of good resources.
And all you have to do is make sure it’s YOUR cookies in the other person’s browser when they start spending money …
Concluding Words of Wisdom
Every day I see people asking interesting and useful questions about Google My Business, crawling, indexing, image search, and so many other things. There are a million useful, interesting topics to learn about in Web marketing.
You’ll get some amazing answers from very smart, experienced people.
There is a lot you can learn from online marketers. But you’re not going to learn how Google’s algorithms work that way. The only surefire way to get that kind of knowledge is to go to work for Google.
If you can’t get a job inside the search giant then you should focus on being expert in the one thing no one else can be an expert in: building the kind of site you think is absolutely amazing.
If you need to hire an SEO specialist to help you get that site into Google and bringing in search referral traffic, remember that it will be easier for your specialist to succeed (and help you succeed) if you avoid all the short cuts, magic formulas, and conspiracy theories and stick to the task of creating that amazing Website.
If you feel the best you can do is copy someone else, I have great news for you: that’s just the beginning of your journey into Web marketing. There are many steps to follow, and you’ll eventually find your own path.
Follow SEO Theory
A confirmation email will be sent to new subscribers AND unsubscribers. Please look for it!