Training Programs

How to Navigate Machine Learning for SEO as an Editor

You can use machine learning for SEO as an excuse not to optimize content, or you can future-proof by following tried-and-true techniques that still work.

While not everybody agrees that Google should control search results as much as they do, we think we can all agree that machine learning for SEO has been a very different ride than the days when it was more manually operated by what we imagine was something like Sergey Brin and Larry Page as the Wizards of Oz.

The thing about machine learning for SEO is that the machine is always going to make mistakes, and in some cases, as it did recently, it can lead to results that affect businesses. In early August, Twitter went aflame when users observed major changes, like Pinterest and parenting blogs ranking highly for health information. A bit of a flub, given that health and other topics like investing have been on Google’s priority list of ranking only reputable sources.

It’s possible that it was just part of BERT (their Ai) trying to diversify search results so that too many of the same website doesn’t show up on page one (a June 2019 update). Or maybe it’s just part of the machine learning process.

Optimizing content for search engines has always been a moving target. There’s no doubt that “writing good content” is the goal, but that statement has never been able to provide those in the publishing industry with any type of guidance. And honestly, it’s not exactly true. Magazine editors have a habit of writing short, catchy headlines that get clicks but have nothing searchable in the title, which as we all know, is the most important element of SEO. So at the very least, keyword research has a place in titles, but also in subtitles, and within the content.

Download a FREE copy of Best Email Subject Lines for Selling Premium Subscriptions and Memberships and discover an extensive list of email subject line frameworks that are consistently proven to sell and boost revenue for publishers.

What Machine Learning for SEO Means for Online Editors

Nothing, mostly. Because you’ll be chewing your fingernails day and night if you try to tackle each new technique and then lose your rank when BERT decides it’s not relevant to its algorithm anymore.

The SEO Scorecard we use and teach editors relies on the most basic signals that Google has always used to help rank content such as:

  • Using a keyword phrase that people actually search for in the title
  • Repeating it in the H2 (subhead) tag, or a natural variation of it
  • Using it within the content naturally, including alt text and other subheadings

There are more that we share with our publishing partners when we train their online editors, and they do evolve along with machine learning for SEO. But these three are the SEO basics that have remained virtually unchanged for over a decade, and BERT is never going decide to rank content for phrases that are irrelevant or not in the content, so these strategies are safe.

Following these best practices doesn’t mean they’ll prevent Google from deciding to drop you for a Pinterest board one day, because that’s the kind of stuff that happens with machine-based learning in SEO. But it puts you in the running to rank and continue being found by people searching for your content.

The next step is really a popularity contest. From everything we’ve seen, Google doesn’t just go ranking articles and content that has never seen the light of day before. This content has an abundance of social media shares and comments.  We recently wrote about Recipe SEO for Food Magazines and shared how structured data and reviews are also a clear factor in ranking food-based content.

We also know that multiple types of media play a part. For example, an instructional post that features a how-to video and multiple images will rank better than a text-based article. These types of articles also get shared more, which adds to their rank-ability.

So while machine learning is affecting SEO for all of us, the big gains and drops we see when there are new algorithm updates are part of the process, and they usually rectify themselves for the better. We can’t tell the storm to go away, but we can build a bigger, more sturdy boat.

Let us help you build an Ark. 

Increasing your website traffic and growing your free email list is key to your ongoing success, as these customer relationships drive all other monetization programs including memberships, sponsorships, and events. Our recommended audience development service package includes us writing, updating, and promoting your SEO posts via email and web, along with managing all aspects of your website email capture program. Our team uses their best practices to leverage your social media traffic, website referrals, and other media relationships. We’ll meet with you monthly to share the ongoing trends using our Audience Development Analytics Suite. Reach out to learn more.

In the comments below, tell us, how you have changed your content writing practices to accommodate constant changes in machine learning?

Download a FREE copy of Best Email Subject Lines for Selling Premium Subscriptions and Memberships and discover an extensive list of email subject line frameworks that are consistently proven to sell and boost revenue for publishers.


Leave a Reply