Is SEO Dead? Yes, the Version Agencies Still Sell You Is

Prefer to watch? The full 12-minute breakdown is above, or browse all our videos.

I’ve been asked the same question on every founder call for six months: did AI kill SEO? The honest answer, the one most people in this space won’t give you, is yes. The version of SEO that most agencies are still selling is dead. But organic search isn’t dead. It’s fractured. It has split across Google, AI assistants and the communities those systems quote, and the companies that understand what actually killed the old model are pulling away from the companies that still haven’t figured it out. In this post, I’ll show you exactly what died, exactly what replaced it, and how to tell which side of the line your company is currently on.

So, Is SEO Dead or Not? The Straight Answer

SEO as a channel is not dead, but the 2022 playbook is: publishing more content, chasing keyword tools and buying any link from a high-DR site. Organic discovery is arguably bigger than ever, spread across Google, AI assistants and communities. In our study of 137 SaaS companies across 1,486 buying queries, 44% of brands ranking in Google’s top 10 were completely invisible to ChatGPT for the same keywords.

That distinction matters more than any headline about SEO’s funeral. The old model was simple. Publish more content, target the right keywords, build any link that comes from an authoritative website, watch traffic grow. Happy days. That loop no longer produces pipeline for most companies, and it’s not because Google stopped working. It’s because the way people search has fundamentally fractured into three distinct layers: classic search results, AI-generated answers, and the community discussions both of them draw on.

Most SEO strategies right now are still only optimising for one of those layers. The companies winning in organic search are deliberately building a blend of all three. So no, don’t cancel your SEO budget. But if your strategy would have looked identical in 2022 (sound familiar?), you should keep reading, because the scoreboard changed and most people haven’t noticed. (If you’re weighing up the terminology itself, we’ve compared the two approaches properly in our GEO vs SEO breakdown.)

What Actually Died (With the Receipts)

Four specific things died: informational traffic as an economic asset, the DR-number link model, guest post farms, and publish-more-content as a strategy. In our experience across client accounts, each one still shows up in agency reports as a win, which is exactly why so many founders feel like their numbers look fine while their pipeline stays flat.

Let me tell you about a call I was on a few weeks ago. A SaaS founder, somewhere around $2-3M ARR, smart guy, sharp on his numbers. He pulled up his agency report and looked almost embarrassed to show me it. On the surface it looked decent. Traffic up 40%. DR up about nine points. Then he goes quiet and says, “But we only got three demos from organic in the whole of the last quarter.”

He wasn’t being cheated. The agency was doing decent work. The strategy was simply misaligned with what he needed, measuring success in a game that stopped mattering. So let’s name the dead pieces properly.

Informational traffic as an economic asset

For a decade, informational traffic was currency. You ranked the guide, harvested the clicks, nurtured the visitors. Across client accounts we see the same shape now: impressions flat or climbing while clicks quietly slide, month after month. The reason is simple. Google’s AI answers absorb the click before it ever reaches your site. The searcher gets their definition or their quick answer inside the results page and moves on. The impression still shows up in Search Console, which is why so many dashboards look healthy. The asset underneath, the actual visit, has already been spent by Google.

The DR-number link model

People have long believed that the more authoritative the website, the better the link. That isn’t strictly true. DR is a third-party metric from Ahrefs, and by itself it means nothing for rankings. It can be manipulated for as little as $5 on platforms like Fiverr, a tactic used precisely to sell backlinks to buyers who only check the number. Chasing links in irrelevant niches, on irrelevant pages, where the DR is the only good thing about the site? Dead.

Guest post farms

Guest posting is probably the most famous link building strategy, but most SaaS editors barely accept guest posts any more. They want their content strategy in-house and product-led. The result is that the majority of “guest posting opportunities” you’ll come across today are link farms, sites that exist purely to sell backlinks. They don’t build authority. Buying links from high-DR guest post farms is a dead strategy, and not one we recommend.

The content trap

Two errors here. First, generic traffic-bait: the “what is a CRM” guide you found in a keyword tool with, say, 10,000 monthly searches. You might rank it, but most of those searchers are students writing research reports, not your buyers. Google has explicitly said, more than once, not to create content purely to capture traffic, and its people-first content guidance spells it out plainly.

Second, bad product-led content: “how to use our CSV export feature” belongs in user documentation, not on your blog. Feature documentation dressed up as marketing, if I’m being polite. Good product-led content is different. It starts from the buyer’s problem, something like “how a 10-person agency cut weekly invoicing from 4 hours to 20 minutes”. Same product, same feature, completely different reader.

What Replaced It?

What replaced the old playbook is a blend: money pages that still earn clicks, citations inside AI answers, and presence in the communities those answers quote. AI systems assemble responses from third-party sources, which means citations are the new rankings. It rewards the same qualities Google has been pushing with E-E-A-T for years, and if you want to go deep on that side of it, Lily Ray is the person to read. Our 137-company study across 1,486 SaaS buying queries in 18 categories found that who gets cited often has little to do with who ranks.

Here’s the single most useful comparison I can give you, two companies in the same category. On paper, Clockify are clearly winning: roughly three times more ranked keywords on Google than Toggl, and about seven times more number one positions. Yet Toggl are cited more often, and in higher positions, for the exact phrases buyers actually type into ChatGPT and Google’s AI answers.

Why? Toggl have more of the right kind of authority. They appear in Forbes Advisor articles, they’re an editor’s choice in PCMag, and there are many Reddit threads organically recommending them. A ton of signals, all saying Toggl is a good option, all from places semantically related to what they want to be known for. ChatGPT and Google are aware of that. Authority hasn’t stopped mattering. The semantic context of your mentions is what matters now.

Communities are now inside the answer path, not adjacent to it. Across the 1,486 B2B SaaS buying queries in our Reddit citation study, Reddit appears somewhere on Google’s AI-enhanced search page 81.6% of the time, rising to 94.1% on bottom-of-funnel “best X” queries, and Google’s AI answer directly quotes a Reddit thread in 12.1% of queries.

Query fan-out explains why: when someone asks Google’s AI “what’s the best time tracking software for my agency?”, that query is broken into a mesh of hidden subqueries behind the scenes: time tracking for agencies, time tracking with invoicing, pricing comparisons, alternatives to the market leaders, reviews, what people on Reddit recommend. The answer you see is the retrieval result of all of them combined. If you only rank for the head term, you’re visible in one strand of a rope. In our fan-out research, we also found YouTube is the most-cited domain in Perplexity’s answers, which should tell you something about how wide “organic” now stretches.

How Can You Tell Which Side of the Line You’re On?

You can diagnose this yourself in about twenty minutes using your own Search Console data and a few AI searches. If your traffic report looks fine but pipeline from organic isn’t growing, that’s the tell. Here are the five signals we check first, drawn straight from the diagnostic in the video.

Recognise three or more of those? It’s time to have a word with your team. If you’d rather have a second pair of eyes on it, our AI visibility audit maps exactly where the gaps are across all three layers.

What Does This Mean for Your SEO Budget in 2026?

It means reallocate, not retreat. The channel still pays when it’s pointed at the right targets: in our Prospeo engagement, we grew organic traffic 1,600% in about a year, from 1,000 to 17,000 monthly visitors, roughly $170,000 in annual traffic value, and the founder credited organic as a big reason they hit $1.5M ARR.

What does the reallocation look like in practice? Fewer posts, better ones, aimed at buyer problems and comparison intent rather than raw volume. Links judged on relevance and semantic context, not a DR threshold, which is the approach behind our SaaS link building work. And a slice of budget for AI visibility itself: tracking where your brand appears in AI answers, building presence on the surfaces those answers draw from (Reddit and forums, YouTube, LinkedIn, podcasts, integration partners), and fixing the citation gaps you find. We’ve written up what that work actually involves in our guide to LLM SEO for SaaS.

Everything in moderation. Don’t be the team that only optimises for AI, and don’t be the team pretending nothing changed. It’s about doing a bit more of everything, deliberately, without spreading yourself too thin. Backlinks and brand mentions still count. They just count for how they amplify you across every surface buyers search, not for a number in a report.

Frequently Asked Questions

Is SEO still relevant in 2026?

Yes, because buying decisions still start with organic discovery, whether that’s a Google search, an AI answer or a Reddit thread. What changed is the shape of the work. Rankings on money pages still convert, and in our 137-company study, AI answers consistently drew on brands with existing search authority. Kevin Indig’s Growth Memo tracks the same shift with independent data, and lands in a similar place.

Is SEO worth it anymore?

Worth it, yes, if it targets buyers rather than traffic. Our Prospeo client grew from 1,000 to 17,000 monthly organic visitors in about a year, around $170,000 in annual traffic value, and the founder credited organic as a big reason they reached $1.5M ARR. Traffic-bait content aimed at keyword tools is what’s no longer worth it.

Did AI kill SEO?

AI killed the click on informational queries, not SEO itself. AI answers are assembled from sources that already rank and get discussed. Across 1,486 SaaS buying queries we studied, Reddit appears on Google’s AI-enhanced search page 81.6% of the time, so the work shifted towards earning citations and community presence.

Is SEO a good career in 2026?

Honestly, yes, but not the 2022 version of it. The people struggling are those who only know keyword tools and link outreach. The skills in demand now span classic search, AI citation analysis, community visibility and content that starts from buyer problems. Broader remit, more interesting work, in my view.

Conclusion: The Scoreboard Changed

So, is SEO dead? The playbook most agencies are still selling is. Chasing meaningless traffic, buying DR, publishing for keyword tools: if that’s your strategy, your reports may look fine while things aren’t so dandy under the surface. But organic discovery itself is bigger and more fragmented than ever, spread across Google, AI assistants and communities, and the data we’ve collected across 137 companies and 1,486 buying queries shows the winners are the ones treating all three layers as one system. The scoreboard changed. Most people haven’t realised it yet. The gap between the companies that have and the companies that haven’t is only going to get bigger.