Author: George Nguyen
Search marketers barely had time to process the potential impact of AI overviews before the internal Google Search documentation leak, and both served as the backdrop for this year’s MozCon in Seattle.
To help you navigate the chaos, stay on the right side of AI usage, and develop workflows that will drive your agency or brand’s success into the future, here are four takeaways for SEO teams from MozCon 2024:
Move past the checklist mentality with context and critical thinking — Lily Ray
Use AI to recommend, not write — Andy Crestodina
Adopt perspective-led content to stand out against consensus-based AI overviews — Bernard Huang
Overcome flawed analytics with directional data reporting — Dana DiTomaso
Move past the checklist mentality with context and critical thinking
“SEO is not a bag of tricks. When you approach SEO that way, more often than not, you’re not going to see any meaningful results.” — Lily Ray, Vice President, SEO Strategy & Research at Amsive
Marketing teams can combat the ‘SEO by checklist’ mentality and achieve better visibility for clients by thinking more critically, holistically, and long-term, said Lily Ray, VP, SEO strategy & research at Amsive, during her MozCon session.
Ray’s advice is a remedy for the increasingly widespread and generic checklist-based SEO resources that some amateur (or just bad) SEOs position as a one-size-fits-all solution.
On top of that, a “vicious cycle of SEO tunnel vision” (illustrated below) that Ray has witnessed over and over during her career managing agency SEO teams is adding to the noise that’s drowning out your client’s actual needs. This overreliance on tools (like ChatGPT) and tactics that worked in the past (regardless of relevance for the current client’s business/situation) can add inertia and, at best, generate lackluster outcomes.
Documented workflows (e.g., checklists) will always be important, especially as a resource for newer SEOs. However, to recommend relevant optimizations that will actually pave the way for business results, you also need to approach each client and their website with fresh eyes—not boilerplate workflows.
To that end, Ray recommends building as much context for the business as possible to develop a holistic SEO roadmap:
Learning the business’s purpose and how it makes money — “The first most important thing that often (for some reason) gets overlooked by many SEOs is learning the website’s purpose and how it makes money,” Ray said. “Why is this part first? Because ultimately, anything else is likely an SEO vanity metric that probably doesn’t matter very much.” SEO can vary dramatically between website categories and business models. Knowing the type of website and the algorithm updates that affect them is a good starting point.
Establish and track KPIs based on revenue-generating activities — This will enable you to establish a baseline and measure your progress. Setting up key events in GA4 to track revenue-generating activities is essential from the outset, Ray said.
Understand the target audience — Look at your various tools and traffic sources to understand where your users come from (referring sites), their age, gender distribution, etc. to discover who your target audience actually is. During her session, Ray recommended tools like SparkToro, BuzzSumo, and SimilarWeb for this purpose. You can also conduct user-first research to identify pain points, potential topics, and other opportunities.
Audit the site’s SEO history — “In 2024, you’re rarely starting from scratch—most of these companies have done some form of SEO and, most of the time, they have some type of skeletons in their closet,” Ray said. Have an ‘SEO therapy session’ with your client and ask them what algorithm updates they were impacted by, whether they’ve ever had manual actions, and if other agencies have worked on the site (if so, ask about their strategies and review their reports). You can also use third-party tools to get ahead of these conversations with your clients, Ray said, recommending Darwin Santos’ GSC Guardian Chrome extension that overlays Google algorithm updates over your GSC charts.
Evaluate the site’s SEO footprint — Find out what’s ranking for your target keywords and get a lay of the digital landscape. Conduct a thorough site: search as well as crawl the site to identify opportunities. This is where your expertise as an SEO comes into play. You need to pair the data from your analysis with your knowledge of SEO (and what’s happening on the SERP) to see who your true SEO competitors are.
Strategize with a group — “If you can get a paid search expert in the room, a tech SEO, a local SEO, a content writer, UX designer, Google algorithm updates specialist, and of course the client, you’re going to get a more holistic understanding of what the business’s needs are and what is the best approach,” Ray said. Success here often depends on getting buy-in for your SEO proposal from a senior stakeholder. But even so, you’ll need to include (and incentivize) every stakeholder whose expertise you’re looking to leverage.
Evaluate the potential risk of AI overviews — SEOs can monitor AI overviews that trigger for their target keywords using a tool like ZipTie, Ray said. This enables you to communicate these risks to your client(s) more proactively.
Develop a holistic, data-driven SEO roadmap — This is the culmination of the steps above. Some rules to keep in mind here are:
Every time you introduce a new SEO tactic, you need to be able to answer “Why?” and connect it to business goals.
If you don’t know the answer to a client’s question, just say so and be honest (and find out later for them)—this can help build trust.
Always be proactive — Inform your client(s) about search updates before they impact them. This can also include new Google product launches, SERP features, spam violations, etc.
Use AI to recommend, not write
“You’ll notice actually, I don’t really use AI to write. I’m using it to do analysis and make minor recommendations.” — Andy Crestodina, Co-founder and CMO at Orbit Media Studios
When it comes to generative AI, much of the buzz focuses on the end goal of creating high-quality, bespoke content in seconds. However, generative AI has largely failed in this regard—especially when compared to human-generated content.
Instead of trying to get ChatGPT to create a ready-to-publish piece of content or copy, you can make more effective use of it by training it to provide relevant recommendations, according to Andy Crestodina, co-founder and CMO at Orbit Media Studios.
While you can always feed it your first-party data, businesses that are starting from scratch can use a variation of Crestodina’s example ChatGPT prompt for B2B businesses:
“Build me a persona of a [job title] at a [industry and/or company size] responsible for [responsibilities/challenges]. This person is looking for [needs/requirements] and they are considering a new [product/service].”
“I want 4 things from AI: [the audience’s] hopes and dreams, their fears and concerns, their emotional triggers, and their decision criteria for selecting a company like mine,” Crestodina said. When analyzing the output (or generative AI outputs in general), Crestodina reminded attendees that the AI also stands for:
Average information — Outputs are generic, making it harder for you to distinguish your brand.
Assume incorrect — Generative AI predicts words; it’s not looking to provide truth. You must double check everything.
Another input — AI is just another tool, which means its potency depends on the person wielding it.
Before moving on to topic brainstorming, Crestodina recommends bolstering the output (based on the example persona prompt above) with some of your own findings about the target audience (as shown below).
Once you’ve vetted your persona, you’re ready to do some brainstorming. Use a prompt like:
“You’re an expert content strategist, skilled at selecting topics that build awareness and trust with a target audience. What information does this person need to do their job well?”
Finally, you can zoom in on particular topics and prompt ChatGPT to provide you with some potential ideas and angles for content.
In the example above, Crestodina told ChatGPT to make the article topics compelling and memorable. “Add those words that work for you, add those words that align with your content strategy,” Crestodina said. “If you want to stay top of mind, I’m telling [the AI] to make it memorable, make it surprising, make it impactful, make it unexpected—whatever the words are that work with your brand, put those in the prompt and copy and paste in that person’s information needs.”
“That, to me, is the right way to use these tools,” he said. “Don’t expect anything good until you train it on your audience, and then talk to it a bit before you get to the point where you’re looking for topics,” Crestodina added, emphasizing that even if only 20% of the outputs are viable, that can still be a worthwhile efficiency gain.
Adopt perspective-led content to stand out against consensus-based AI overviews
“AI overviews are designed to answer the consensus queries that everybody is more or less agreeing upon. And what we’re seeing then, is perhaps the biggest cannibalization of keywords and queries that we’ve ever seen in Google Search by Google. If they’re taking a look at the ‘What is,’ and the ‘How-to’s,’ [Google is] just saying, ‘Yeah, everybody’s more or less saying the same thing—so, we’ll say it.” — Bernard Huang, Co-founder at Clearscope
If Google reserves the top of the SERP for its AI overviews, then what’s left for brands and SEOs? During his session, Bernard Huang, co-founder at Clearscope, reconciled Google’s guidance on experience-rich content with its most recent and controversial SERP feature, pointing to information gain and ‘perspective-led content’ as the type of content that Google now rewards.
‘Consensus topics/queries’ (mentioned above) represent agreement on an established topic and are therefore more likely to trigger an AI overview that disincentivizes clicks. Information gain, on the other hand, adds new entities to Google’s knowledge graph, enabling it to better understand (and provide results for) new queries, potentially avoiding AI overviews as well (due to a lack of consensus on the topic).
To that end, “instead of this ‘skyscraper’-type content that we’ve all been encouraged to do over the last 5-10 years, what we need to really be thinking about is perspective-led content,” Huang said, urging SEOs to switch from targeting keywords to instead targeting perspectives. Reddit’s recent prevalence in SERPs seems to support this interpretation.
Huang recommends that SEOs and content creators view their topic through the perspectives that users are likely to engage with, such as:
Things I wish I knew (e.g, “7 things I wish I knew before starting the keto diet”)
Can anyone learn XYZ? (e.g., “Can anyone learn to code?”)
Should you XYZ? (e.g., “Should you try the keto diet?”)
For all skill levels (e.g., “Marathon training plans for every runner”)
Time urgency (e.g., “How to train for a marathon in a short time period”)
Without XYZ (e.g., “10 ways to control high blood pressure without medication”)
DIY (e.g., “10 do-it-yourself SEO tips to save money”)
And, for sustained performance, Huang advises that brands and SEOs be vigilant in monitoring and refreshing their content—this best practice is even more important as new AI-generated content enters the ecosystem.
Overcome flawed analytics with directional data reporting
“[Analytics is] actually getting worse: Consent management, ad blockers, Safari—lots of things are making it very, very difficult for us to get any sort of accuracy whatsoever when it comes to our data.” — Dana DiTomaso, Founder and Lead Instructor at Kick Point Playbook
Analytics has always had its limitations, like web browsers, for example: “If I go to your website on my phone and I go to your website and my desktop computer, I am two people,” said Dana DiTomaso, founder and lead instructor at Kick Point Playbook. “There is nothing in GA4 that is going to tie these people together.”
To make matters worse, shifting consumer privacy preferences (and how each tech giant seems to want to approach the issue) has only added more caveats and nuance to work around and convey to clients.
With these challenges in mind, DiTomaso presented attendees with two paths: “We could continue to pretend that we can track everything—wrap ourselves up in a cozy blanket of lies —or we can accept the truth of the situation … So, instead of trying to be like, ‘This totally happened and this is how many times it happened,’ how about we embrace ‘just good enough?’”
‘Accepting the truth of the situation’ means acknowledging that the data is far from perfect while still referencing it for strategic decision making and reporting. Since the data is not 100% accurate, DiTomaso recommends using ‘directional data’ in your reporting instead of aiming for precision.
Directional data points at what you need to know and helps you answer directional questions, like:
Which landing page is doing a better job of driving signups? By providing metrics as percentages and including bar graphs (as shown in the example below), it’s far easier to discern which pages contribute the most to conversions (against their respective shares of total sessions).
Is our organic traffic increasing or decreasing? Similar to the tactic above, you can also use percentages to convey your performance over a time. And in most cases, it’s okay to round off percentages to simplify the report for stakeholders.
Is this video encouraging people to buy our products? For this example, DiTomaso created two audiences in GA4 (people who had a video start event and people who have never had a video start event) and compared their conversion rates (key events).
DiTomaso also introduced ‘effectiveness’ as a directional reporting concept. “‘Effectiveness is a measure of how good your marketing is at hitting your goals,” she said, explaining, “It’s based on two metrics—they may not even be metrics in the same system; you just see how they relate to each other and that’s it—that’s your effectiveness.”
For example, you could evaluate device sessions against conversions (as shown below). “Do the slices match up? If they don’t, something is wrong,” DiTomaso said, adding that this particular example could be used to solicit buy-in for a mobile site redesign.
This type of analysis can help you answer directional questions like:
Are your emails spurring action?
Did those technical SEO fixes improve revenue?
Did we reduce churn with our new learning series?
Stay up to date with live SEO events on the Wix SEO Learning Hub
This year, MozCon featured over 24 expert speakers. In addition to this recap, I encourage you to check out their sessions; these were some of my favorites:
On the Hunt for Hidden Gems: Perspectives on UGC in the SERP — Crystal Carter
The Future of Search — Rand Fishkin & Dr. Pete Meyers
SEO Mind Games: Decode Searcher Bias for Content Success — Garrett Sussman
The Power of Emotion: How To Create Content That (Actually) Converts — Talia Wolf
Build a Story-Driven Marketing Machine — Joel Klettke
Multimedia Marvels - A Symphony of Audio and Visual Marketing Narratives — Azeem Ahmad
For their full decks, see MozCon’s schedule page.
If you enjoyed this recap, subscribe to Searchlight, the Wix SEO Learning Hub’s monthly newsletter, to find out about our most recent resources and event coverage. And, check out our previous event coverage from this year:
George Nguyen is the Director of SEO Editorial at Wix. He creates content to help users and marketers better understand how search works. He was formerly a search news journalist and is known to speak at the occasional industry event.