Predictions and Hypotheses from 23 Years in SEO
Google announced a new signal for limiting the appearance of low-quality / not authoritative content in organic search. This signal focuses on content that doesn’t “help”.
- Announced: August 18, 2022
- Rollout: Begins the week of August 22, 2022 (lasting up to 2-weeks)
- Searchers Impacted: English users, globally
- Goal: "Make it easier for people to find helpful content made by, and for, people"
- Industries Specified: Online Education, Arts and Entertainment, Shopping, and Tech
- 'Helpful' Signal: Site-wide and weighted, Google's process is entirely automated
- 'Unhelpful' Classification: Content can still rank well if other signals identify it as helpful/relevant
Unhelpful content could be everything from a complex topic written by an individual who doesn’t have strong expertise in that topic, but follows outdated “SEO Rules” or not even written by a person at all.
Scenario 1: "Hey have this sophomore intern write up this best CRM guide, make sure to keep it to 1500 words with 1 H1 and 2 H2’s."
Scenario 2: GPT-3 generated content – a language model that uses deep learning to auto-generate human-like text.
While that content can read well, does it truly offer expertise? Imagine if we allowed GPT-3 to write this article you're reading now (which it could do and it would sound great) vs my expertise of 23+ years of Search (started in August of 1999). Which would you follow the advice of? Which content is more likely to be helpful to the user?
Google’s advice for this update echoes what they've been saying for years:
- Focus on people-first content
- Avoid creating content with the primary objective being search engines
This advice is often so broad that it leaves content creators asking more questions rather than getting answers - but that is what we’re dealing with.
When I think back to updates, especially the old ones like Penguin and Panda, Seer always embraced those because it validated the stuff we were trying to push. These updates got past the “solving for users & writing helpful content” leading to a failing SEO campaign instead of being rewarded.
"My mantra, I’d rather be thanked than ranked."
Here is a link to my video on this from Mozcon 2017. It grounds us in writing helpful answers to problems as a way to get rankings, instead of putting “get rankings” first.
We all know that once we get ranked, too many of us think the job to be done is over. But shouldn’t we be constantly evaluating that content and tweaking it to ensure it stays helpful?
New information is entering our world all the time and that creates a SERP that is constantly changing. If you don't keep up, you could end up having a rapper come in and cost you a few million $$s in paid spend, pulling your organic rankings down with it. Savage.
Wil's Prediction: What Queries Will Be Impacted
Some queries are simply “fact” driven and some are built to help you make a decision or “expertise” driven. It's the latter that I believe will be affected the most - those queries where folks want to research a level deeper.
Example: 'What is a ribeye steak' is very different from 'what happens when I dry age a ribeye for 1 hour vs 1 day vs 3 days'.
Simple "Fact" Driven Query: Searchers are going to read that first meta description and most likely have what they need:
Make a Decision / 'Expertise' Driven Query: Searchers are going to have to dig into the content to get the answer they need. It's going to take more than a 5-second scan. And for this one specifically, you can see Google knows most users want to watch a video to better understand this question:
Wil's Prediction: Who Gets Impacted
My bet, publishers should watch out.
We'll validate this with a large study as the update rolls out.
I'm betting these are the folks that get hit the hardest:
- Publishers focused on a very broad range of topics (CNET, Forbes, etc.)
- Sites that exist to collect and monetize organic search traffic, without providing unique value
- Sites that leverage content created with tools like Jasper or Copymatic
There is a business case for looking at people who aren’t your direct competitors and our big data approach to SEO helps us find the head-scratching moments when someone we don’t expect is outranking the sites in the space with topical authority.
Think you only have 5 to 10 competitors? Then you're going to want to watch this.
Wil's Recommendations: What You Can Do
Google’s announcement gives us extensive clues to signal what content may get knocked down, how greatly our websites may be impacted overall, and how we can move forward.
I’ll go through them one by one – breaking down Google's advice, layering in my hypotheses, providing examples, and any takes I have based on the data I’ve reviewed combined with my 23 years of experience in the SEO industry.
We have 3 basic areas of advice to keep your content high in the SERPs with the new Helpful Content Update:
- Stay in your lane: write about topics that pertain to your business, that you’re an expert on, and are for your audience
- Answer the searcher’s question: provide an expert and unique perspective along with answers to likely follow-up questions
- Don’t hack the algorithm: aim for content lengths that satisfy the reader’s query
These are a roll-up of Google's advice from "What creators should know about Google's helpful content update."
#1 Stay in Your Lane
Here’s what Google qualifies as “staying in your lane”:
- Do you have an existing or intended audience for your business or site that would find the content useful if they came directly to you?
- Does your content clearly demonstrate first-hand expertise / a depth of knowledge (ex. from having actually used a product or service, or visiting a place)?
- Does your site have a primary purpose or focus?
Here’s what Google qualifies as “not staying in your lane”:
- Are you producing lots of content on different topics in hopes that some of it might perform well in search results?
- Are you writing about things simply because they seem to be trending and not because you'd write about them otherwise for your existing audience?
- Did you decide to enter some niche topic area without any real expertise, but instead mainly because you thought you'd get search traffic?
My Hypothesis: Not staying in your lane will greatly affect publishers like Forbes, CNet, Mashable, etc.
My Advice: Watch content velocity, are you suddenly showing up with “lots” of content, across various themes off of your main topic? If so, keep an eye out.
Example: Forbes.com ranked 2nd for “best truck tires,” even though they are a self-described “global media company, focusing on business, investing, technology, entrepreneurship, leadership, and lifestyle.”
Forbes ranked higher than 3 other sites clearly focused primarily on Automobiles.
When you run Forbes’ homepage meta description through Google Cloud’s Natural Language API demo tool - you’ll see the categories Forbes focuses on.
We just did this quickly, but if anyone has a link to a tool that helps you get all entities of a site based on a few URLs - please hit me up on Twitter.
While if we take Car and Driver’s meta description, we get this:
I believe sites like Forbes will rank lower on topics that don’t directly relate to their core mission: business. They may even slip on business terms since this is a weighted and sitewide signal.
What You Should Do Instead:
Understand your primary business objectives - what matters to your c-suite. Map those back to the audiences who they serve. Know. Your. Audience.
Try These Resources:
- How to Find Your Target Audience (HubSpot Blog)
- The Guide to Designing a Content Strategy (Brain Labs)
- You Don't Need a Buyer Persona. You Need Multiple Audience Personas (SparkToro Blog)
- Business Acumen 101: 5 Concepts Every Marketer Needs to Know (Seer Blog)
#2 Answer Your Searcher's Question
Sounds simple, right? Not necessarily. You need expertise, research, and time - it's an investment.
Here’s what Google qualifies as “answering your searcher's question”:
- After reading your content, will someone leave feeling they've learned enough about a topic to help achieve their goal?
- Will someone reading your content leave feeling like they've had a satisfying experience?
Here’s what Google qualifies as “not answering your searcher's question”:
- Are you using extensive automation to produce content on many topics?
- Are you mainly summarizing what others have to say without adding much value?
- Does your content leave readers feeling like they need to search again to get better information from other sources?
- Does your content promise to answer a question that actually has no answer, such as suggesting there's a release date for a product, movie, or TV show when one isn't confirmed?
My Hypothesis: This affects sites that bought “cheap content” or use algorithmic approaches of entity matching to produce content based on what is already ranking on Google. If you jumped on the GTP-3 train when it took off in 2020, started creating content with tools like Jasper or Copymatic, and didn’t augment it with expertise - you could be identified, let's see.
My Advice: There are no shortcuts to creating great content at scale, but there are frameworks and solutions that make it easier for all parties. The "What You Should Do Instead" section below has some things we’ve successfully done with busy teams that have aggressive goals.
Example: If you told a tool that auto generates content using GPT-3 to write a page on a topic and told me to write on the same topic – my page would smoke GPT-3 to the human eye. But up until now, Google wasn’t able to catch that.
What You Should Do Instead:
- Understand your customers - Talk to them, study their problems, their triggers to finding solutions, and their interactions with your brand.
- Talk to your customers. Yes, it takes time. But it's worth it.
- Understand the questions in People Also Ask boxes. Notice how Google talks about having to search after a search to find answers? How else will you know what searchers ask for next?
- Map those out into your content strategy - answering the burning questions you know your existing audience is bound to have to solve.
- Talk to the experts
- Find time with someone on the sales or customer service team, and record and transcribe your conversation.
- Years ago we had a client who only had time to chat on their drive home. We took it, and we were able to collect great information from an individual with deep subject matter expertise and turn it into the content.
- Experts have opinions, ask them what they feel about trending topic X or Y vs just regurgitating facts back.
- Use trends, PAA clues, and news to drive your questions and ask them what they think.
- Hire experts and create logical/efficient workflows or risk penalties. The choice is yours but if this thing rolls out and takes hold, we’ll need to find a way to balance scale and helpfulness in our content. This takes more buy-in, but our clients who are most successful in creating content at scale have in-house resources that they pair with freelancers.
- Use the resources that already exist. Does your sales team have old webinar content that explains their product and solution? What about sales collateral? When you’re in a pinch with resourcing, repurposing other types of media can be a great strategy to get ahead.
Try These Resources:
- Optimizing for People: Why User Research Matters for Marketers (Seer Blog)
- Audience Research in Marketing: Start with the Data You Have (Seer Blog)
- Content Marketing Remix: Turn One Post Into A Deck, Podcast & More (Ross Simmonds)
- GPT-3 & SEO: What You Should Know (Seer Blog)
- Using People Also Ask At Scale to Write Content (Seer YouTube)
- Free Template for Optimizing Landing Pages Using PAA NGrams (Seer Blog and DIY Template)
#3 Reverse Engineering the Algorithm Won’t Help You Write Expert Content
Is anyone still really doing this? Aren't you tired of chasing the algorithm? It's much easier to chase your user. That's who Google is chasing, so you're just skipping the middle man by focusing on people first.
Here’s what Google qualifies as “not reverse engineering their algorithm”:
- Are you writing content for your user and not for the search engines?
- Are you keeping in mind our guidance for core updates and for product reviews?
Here’s what Google qualifies as “reverse engineering their algorithm”:
- Are you writing to a particular word count because you've heard or read that Google has a preferred word count?
- Are you writing content made primarily to exploit search trends vs actually helping humans?
Hypothesis: If you’ve been doing this, you’ve probably already experienced big dings over the past few algorithm rollouts from Google.
Example: You should already know if your site will be impacted by this, as it’s a very intentional tactic to create misleading content. This isn’t something you can accidentally fall victim to.
What You Should Do Instead: So if you want to stop chasing Google, start chasing your customers. Sit with them, ask them questions, and get feedback on your content…imagine if companies started having a content counsel where they asked customers to review content for helpfulness before it launched. Asking what is missing, what else would you have to Google to find it…that could be an interesting direction.
Try These Resources:
- How to Create Content that Converts (Siege Media)
- Everything we've already listed above
What It All Comes Down To
Google directly called out that “sites with relatively high amounts of unhelpful content overall are less likely to perform well in Search.”
As a result, their advice is to get rid of that unhelpful content to help increase the rankings of your other content.
What’s Next for the Helpful Content Update?
Because we’ve been pushing the envelope on big data and SEO for years we have lots of ways to slice and dice millions of keywords and hundreds of millions of ranking URLs, we’re going to go DEEP into the vault, and we'll share everything we can about winners and losers.
For example, we've started looking back over 12 months at which publishers built the most “new pages” of content, so we already know who to have our eye on.
To track the impact of this update on your website health and performance, make a copy of this Google Data Studio template.