Scared but Optimistic: Why Automation Is Killing Entry-Level SEO (But Not SEO)

In this week’s episode of Voices of Search, we spoke with Patrick Stox from Ahrefs about the uncomfortable truth nobody wants to say out loud: we’re scared. Not about SEO dying—but about what happens when one person can do the work of ten.

Patrick, who’s spent over 20 years in the industry and led technical SEO at IBM before joining Ahrefs, opened up about automation’s societal impacts, why the pipeline for new SEO talent is at risk, and how he’s building tools that let you fix 95% of technical SEO problems with a single click. He also revealed why Ahrefs is shifting from top-of-funnel content, how the company thinks about brand visibility in AI responses, and what it means to build an SEO team when job titles matter less than having people with ideas.

Key Takeaways From this Episode:

• Automation will eliminate most entry-level SEO work within years, creating a talent pipeline crisis where interns who once rewrote title tags have no path into the industry.
• SEO is shifting from in-house optimization to external visibility work across Reddit, influencer partnerships, and brand reputation because AI systems prioritize consensus over single-source claims.
• Current professionals are probably fine, but newer folks face a compressed career ladder where you need domain knowledge and creative ideas to compete with tools that can execute basics instantly.
• Ahrefs is building real-time auditing with one-click fixes that bypass dev teams entirely, using AI hype to finally sell edge workers and middleware that’s been technically possible since 2004.
• Traffic from AI search is still under 0.5% while Google remains 40%, but Ahrefs is hedging bets by shifting toward bottom-of-funnel content and putting real faces behind their brand.

Why “Scared” Is the Honest Answer

When asked how he feels about the state of the industry, Patrick didn’t sugarcoat it: “Scared.”

“The reality is, not quite yet, but within a couple more years, you’ll probably need one person for what 10 people were doing before. And the societal impacts of that are scary to me,” Patrick explained. “Will people work less, and we have the same number of people? Will there just be one person for every 10? Will there be riots in the streets, or will we make the Star Trek utopia?”

This isn’t about SEO disappearing. Patrick believes current professionals will be fine. “We probably have a good run rate. But newer folks coming in—what do we have SEO interns doing? Rewriting title tags and meta descriptions, and adding internal links. Basic things.”

Those basic things are exactly what AI automates best. “AI is really good at that. So what do we do now? Jump into a mid-career, mid-level? I think a lot of companies aren’t going to be willing to take that kind of risk.”

The Pipeline Problem

The real concern is the talent pipeline. “I think the people that are already in, that have experience, that are willing to adapt to these new systems are going to be fine. Those that are not willing to adapt maybe not so much. But the pipeline for new talent is at risk of drying up.”

Patrick drew a comparison to law. “A lawyer that’s 70 years old has all these relationships—he’s fine. But the paralegals, new lawyers getting into it, you’re just going to need fewer people, and the pipeline is going to be difficult.”

This creates a paradox: automation makes experienced practitioners more productive, but eliminates the entry points where they built that experience in the first place.

The Shift From Internal to External SEO

Patrick described a fundamental strategy change at Ahrefs that reflects broader industry shifts:

“Ahrefs’ blog was top-of-funnel informational, but we had people read and then collapse the funnel. AI is taking that. People aren’t going to go read the blog because it was cited. No one is clicking in AI search. They’re going to ask five more questions to AI search before they click through.”

The response? Move beyond owned properties. “We never did good on things like comparison pages or managing how we’re talked about online. A lot of what we’re doing around those tactics are changing. Now it’s influencer and outreach.”

The Multi-Site Consensus Strategy

Ahrefs was even considered a controversial tactic. “We’ve talked about leveraging some of the other websites—Tim’s old site, Glenn Allsop, and a few others. Do we just publish the same content on five, six sites? We’re more likely to get the AI to say what we want to say if we do that.”

But Patrick doesn’t love the approach. “I don’t particularly like spamming the web like that. But it’s what’s working now, at least for a company our size. Someone could easily abuse that—hundreds, thousands of websites all saying the same thing and likely get the AI to say what you want.”

This reflects how AI systems work: they repeat what they see consensus on. “These systems basically just repeat what they see. If it’s said on the internet, that’s what they’re going to say.”

The Inevitable Consolidation

Patrick predicts AI platforms will scale back citation volume. “Right now, as many citations are being thrown, no one is clicking. Website owners are going to revolt. SEOs are going to revolt. The numbers don’t look good.”

He’s publishing research showing the traffic impact. “I have a few blogs looking at ChatGPT versus Google and AI search versus traditional search, even the click-through rate and modeling all that. Website traffic is going to go away.”

The response will be consolidation. “I think they’ll take these other signals and condense them down and say instead of citing these 80-some websites, let me just cite these five that I know and trust.”

What Signals Will Matter

Traditional SEO signals will likely influence which sites survive consolidation. “Things like links will probably make a difference, or user signals. Was this site getting traffic? Is this a brand people are searching for? How have people been interacting with this website? Google has all that data from browsers and Android.”

For OpenAI without those signals, Patrick sees challenges. “I don’t think they have anything like PageRank or a link index. It might be augmented from other systems. They may get this data from a third party. They had a partnership with Bing for a while, but I don’t think that’s a thing anymore.”

The Reddit Question

On Reddit’s dominance in AI training, Patrick is cautiously optimistic but realistic. “It’s being spammed to death right now. If the data scientists do a good job cleaning the data, it’s trustworthy. But honestly, if they look at it overall, it’s getting less trustworthy.”

He predicts decline but not disappearance. “I think we’ll see somewhat of a decline, but it’s not going away because it’s still great content.”

The pattern repeats across platforms. “Digg just relaunched as a Reddit competitor. It’s already being spammed to death by SEOs. I look at it, and I’m like, ‘No, Digg is dying again already.'”

Trust Moving Back to Websites

Patrick sees early signs of models prioritizing official sources over third-party claims. “We were seeing data poisoning where people would make up that Ahrefs didn’t have a certain feature. Do you believe that, or do you go to Ahrefs and check? That’s what they’re doing more now—actually more likely to trust the website itself than third-party sources.”

This wasn’t always the case. “That wasn’t a thing for a while. I do think we’ll probably see more of that. Why am I trusting Medium and Reddit and 50 bajillion blogs no one ever heard of versus let’s just go to the source?”

Building the Technical SEO Dream: One-Click Fixes

Patrick’s most exciting work at Ahrefs combines real-time auditing with instant implementation through edge workers.

“We have an almost real-time auditor. We get the index data, all the page changes. If content’s created, you delete something, you redirect something, it tells us. We see that immediately and can be like, ‘Oh, you changed the canonical tag. You got that wrong. Did you mean to do that?’”

The Edge SEO System

The fix mechanism uses edge workers—serverless technology that intercepts requests before they reach servers. “We have this edge SEO system that lets you change things before it’s sent to servers or bots. If the canonical’s wrong, we can be like, ‘Okay, change it back to how it was.’ Boom, it’s fixed. You don’t have to go into your CMS.”

This solves Patrick’s longtime frustration. “Instead of logging into WordPress and finding where this section of text was, you just go, ‘I want that link, I want that link, I want that link.’ Done. Click yes, and it implements for you.”

The technology isn’t new. “Edge workers have been around since 2016. Cloudflare launched Workers in 2017. This serverless technology has been a thing. Even before that, you could use middleware, a reverse proxy. Stefan Spencer had something like this in 2004.”

Why Now?

What changed? The ability to sell it. “Most people have never heard of this tech. It’s been doable, but devs push back. Infrastructure teams are like, ‘We’re not putting a middleware system in there.'”

AI hype provides the opening. “Now it’s AI that fixes everything on the website. Click. Yes. Execs are going to eat that up. Dev teams are going to have a hard time arguing against that. I think it will finally work and see mainstream adoption this year.”

Patrick’s goal is ambitious. “By the end of this year, I want to solve like 95% of technical SEO for websites. You see the problem, you want us to fix this, you click yes, done.”

Where SEO Time Should Actually Go

With automation handling technical cleanup, where should practitioners focus?

“Probably still on your content. That’s where it should have been spent anyway,” Patrick said. “There are a few things in tech SEO that matter, but for the most part, if you’re indexed, you’re fine. Creating better content, figuring out how to do that, getting more resources there, more insights.”

Getting Expert Knowledge at Scale

Patrick sees AI enabling better content through expert extraction. “We have this thing in Slack now, and I’m like, could I have that in people’s Slack? If you know about this topic, let me use AI and generate five questions. They can fill this out, and I can use their info.”

The possibilities expand from there:

• Ingest internal and external knowledge bases
• Have AI conduct 5-10 minute interviews with pre-generated questions
• Process sales team call transcripts for common questions and complaints
• Extract what customers are asking to guide content strategy

“Before, you talk to a person, and they forget five things, and you get a fraction of the info. Now I think you can get more out of that and make better content.”

The Content Prioritization Debate

Patrick is shifting focus. “I’m moving more towards end-of-funnel stuff. Ahrefs blog was a tremendous top-of-funnel. Our YouTube is actually still a tremendous top-of-funnel. Video content still works incredibly well.”

But there has to be a balance. “Google is still our number one referral source. We have data from web analytics—Google search is still like 40% of traffic. I think all the AI search systems together are under half a percent still.”

His advice: “There has to be some mix right now. If you’re skewed in one direction or another, it’s not the best mix. You don’t want to completely not be present for the impact you can have in brand recognition and click-through rates when they hit later stages.”

Putting Faces Behind the Brand

Ahrefs deliberately shifted to personal branding alongside company branding.

“You’ve seen the shift on social media. Instead of just the brand, you have faces. You have me, Tim, Ryan Law, and Glenn Allsop. You put real people behind it. Sam O for YouTube videos—every SEO knows who Sam O is because he’s the face of Ahrefs on YouTube.”

The reasoning is fundamental. “People connect to people, not necessarily to brands. If I see something quoted from Ahrefs, I may be like, ‘Yeah, whatever.’ But if I see, ‘Oh, that’s from Ryan Law,’ I’m like, ‘Oh, I love his stuff. Let me go read that.’ Now people will actually click.”

This human element becomes more critical as automation increases. “There is still a human connection, interaction, and reputation aspect that’s very difficult to get away from. Having that at the center of strategy and how it’s being presented is key.”

Building Networks Over Blogging

Patrick was asked if he’d start a new company with SEO. His answer was surprising: “No, I would go to my network, have them help me advertise it. I’m lucky enough to have built a network—a micro niche influencer. I have a pretty passionate network that likes what I do, generally. They would be how I would try and start, not go write a bunch of blogs.”

He might still create a video, though. “I might go make a bunch of YouTube videos or shorts, TikTok, Instagram reels, that kind of thing.”

This reflects a broader shift: relationships and audience matter more than owned content in a world where AI intermediates discovery.

The Dream SEO Team

When asked how he’d build an in-house SEO team from scratch, Patrick focused on character over credentials.

“You’re going to need someone for technical, at least one person for content, probably multiple. Video is going to be a core part of the team. But job titles don’t matter as much as I just want creative people who are willing to learn and try things and do cool things.”

The differentiation matters more than volume. “You could have a bunch of people for content, but if they don’t have ways to differentiate their content from the 50 other companies that do the same thing, what’s the point? If you can have people that stand out, that you can put a name and face behind, that’s going to go a lot further than a generic person behind a company name.”

Ideas Over Skills

Patrick’s key hiring criterion: “I want people with ideas. You don’t need a thousand devs anymore. You need someone with domain knowledge who can come up with a thousand ideas.”

He acknowledges the challenge. “That’s one of the harder things to find and hire for. Anyone can give you that checklist of years of experience and specializations, but really judging and seeing through the experience to the character of the person is the most valuable, but by far the hardest to gauge. You can’t just tell someone what the criteria are. It’s more of a feel, something softer.”

What Hasn’t Changed: Be a Sponge

Despite all the disruption, Patrick’s career advice for new SEOs remains consistent with what he’d say 15-20 years ago: “Be a sponge. Things are changing. You have a lot to learn. You’re probably going to learn some wrong things. So just be adaptable. Things you learn now may not be correct or the way they will be in five years.”

The industry’s willingness to share knowledge remains a strength. “That’s one of my favorite things about the industry. For the most part, there’s an overall willingness to share information and share knowledge. You don’t have as much of someone willing to take the time to explain something or lend a hand where they may not have immediate business value in doing so.”

Patrick calls it “abundance mindset. There’s plenty of work for everyone. More than enough work. There’s no reason not to do it.”

He also found personal benefit. “I learned a lot more once I started sharing, writing, and speaking. I would go down deeper rabbit holes, do more testing. That brought my skill set up. So it helped me by helping others, which is awesome.”

Naming the Future

In what we’ll call this new era of search, Patrick thinks we’re overthinking it. “We got another year or two, and then it’s just all search again.”

His bet: “SEO.” Simple. Direct. No new acronyms needed.

Tyson suggested “search everywhere optimization” might work, and Patrick agreed it’s “pretty true” given the expansion beyond owned websites.

But ultimately, “It’s all still search. I’m making the distinction between AI search and traditional search. We got another year or two, and then it’s just all search again.”

Staying Motivated After 20 Years

What keeps Patrick engaged after two decades? “We have a lot of changes now, and things were kind of the same for a long time. You just did the same things. So it’s more interesting now.”

But the deeper motivation is contribution. “A big part of my motivation is just giving back. The industry gave me so much. I just want to help people now. I want to make your job easier, your life easier. I want new people to improve their skills and do really cool, fun, great things.”

SEOs Are Probably Fine (But Entry-Level Isn’t)

Patrick’s “scared” opening wasn’t about personal career concerns. Current practitioners with experience and a willingness to adapt will thrive. Automation amplifies their impact rather than replacing it.

The fear is structural: what happens to an industry when the bottom rungs of the career ladder disappear? How do you build expertise when the tasks that built yours no longer exist?

Patrick doesn’t have perfect answers. Neither does anyone else. But he’s building tools that bet on experienced practitioners becoming more powerful, not obsolete.

The question now isn’t whether SEO survives. It’s whether we can build new pathways into an industry that’s removing the old ones.

Voices of Search is a daily SEO and content marketing podcast hosted by Jordan Keone and Tyson Stockton. The show delivers actionable strategies and data-driven insights to help marketers navigate the ever-evolving world of search engine optimization and content marketing. New episodes air weekly, covering everything from technical SEO to AI discovery, featuring industry leaders and practitioners sharing real-world frameworks and proven tactics.

Subscribe to Voices of Search on Apple Podcasts, Spotify, or your favorite podcast platform. Follow Previsible on LinkedIn for updates and subscribe to the VOS YouTube channel for video episodes and clips. You can also visit the official VOS site to explore the full episode archive and submit your SEO questions for future episodes.