Marketing Spark
All episodes
From Marketing Spark · Jun 18, 2024 · Sam Dunning

AI Overview SEO: What Still Works for B2B After the Google Leak

I woke up to the Google leak the same morning Sam Dunning did — Rand Fishkin's post in my feed, coffee in hand, the realization that everything Google had publicly denied for years was sitting there in plain text. A few weeks later, AI Overviews started eating the top of the funnel. If you run a B2B SaaS company and you're trying to figure out what ai overview seo even means now, this is what Sam and I worked through on Marketing Spark.

I woke up to the Google leak the same morning Sam Dunning did — Rand Fishkin's post in my feed, coffee in hand, the realization that everything Google had publicly denied for years was sitting there in plain text. A few weeks later, AI Overviews started eating the top of the funnel. If you run a B2B SaaS company and you're trying to figure out what ai overview seo even means now, this is what Sam and I worked through on Marketing Spark.

Drawn from Marketing Spark Episode 33 with Sam Dunning, founder of Breaking B2B.

The leak confirmed what good SEOs already knew

Sam's reaction to the leak was the same as mine: surprise, then a shrug. Links matter. Brand matters. Click data matters. Google had spent a decade brushing those off in public Q&As and then quietly ranked sites on exactly those signals. "It was reassuring for a lot of folks that have been doing SEO for a little while to see that some of these things are actually ringing true," Sam said. If you're a B2B founder, the leak isn't an action item. It's not a 200-page document you need to read this weekend.

The thing that actually changes your job is sitting in the SERP itself: Google AI Overviews. When a buyer searches "what is web design" or "how to build a SaaS pricing page," they no longer scroll past ten blue links. They get a synthesized answer, two or three supporting links, and a reason to never click anything. The BBC reported on sites that lost 90 to 99 percent of traffic almost overnight — some of them shut down. That's the part to take seriously. The leak is a story. AI Overviews are a P&L event.

Sam's diagnosis is uncomfortable for most B2B marketing teams: the content you've been publishing to chase top-of-funnel keywords — "what is X," "how to do Y," "best ways to Z" — is exactly the content AI Overviews replace. If your SEO strategy was built on that, you don't have an SEO strategy anymore. You have a content archive.

Why most B2B teams get ai overview seo wrong

Sam laid out the failure mode I've watched dozens of $5M to $20M SaaS companies fall into. The exec team says "we're not getting demos from organic, can you do some SEO?" The marketer — already juggling 102 other tasks — does a quick keyword pass, runs a blog post through ChatGPT, hits publish, and forgets it. Three months later: "Great, the post got traffic. How many demos?" "You didn't ask for demos. You asked for SEO."

SEO sits at the bottom of most B2B budget stacks. Paid, events, sponsorships, outbound, partnerships — all get fed first. SEO gets the scraps. So teams chase whatever's easy: high-volume top-of-funnel keywords that produce a traffic chart that looks like a hockey stick and a pipeline number that looks like flatline. AI Overviews just turned that strategy from inefficient to actively pointless. The robot answer at the top of the page now gets the click your blog post used to get.

If you're spending budget on top-of-funnel SEO right now, Sam's advice is direct: stop. The volume is being eaten. The intent was never qualified to begin with. You can't keyword-research your way out of this. You need to rebuild the program around buyers who are ready to talk to sales.

Build the spreadsheet, then go after bottom of funnel

Sam walked me through the exercise he ran when he started Breaking B2B four months ago. It's a Google Sheet. That's the tool. Three columns of inputs:

  • Offers. What are the main things you sell? Be specific. "SaaS" isn't an offer. "Parallel dialer for outbound SDR teams" is.
  • Industries. Which sectors have you sold into where the deal closed quickly and the customer renewed? Not who you wish you could sell to — who has the expensive problem and the cash to fix it.
  • Competitors. Pull your top ten from G2, Capterra, or just from sales call notes. Calendly, Chili Piper, Revenue Hero, HubSpot meetings — the usual suspects in your category.

Then add a fourth column: jobs-to-be-done. The struggling moments. A sales rep dialing manually from their cell phone at ten dials an hour thinks, "there has to be a faster way to reach decision makers," and that exact sentence is a keyword. "How to make faster dials with my cell phone" is the search. Your article introduces the parallel dialer category and shows how SDR teams hit 100 dials an hour. That's bottom of funnel seo done right — it captures someone who is already mid-job and one click from a sales conversation.

Run those keywords through Ahrefs. You're looking for modest volume with high intent, not 10,000-search-a-month vanity terms. Industry + solution. Best + tool. Competitor + alternatives. Job-to-be-done queries. "Exhaust the bottom of the funnel," Sam said. Most B2B companies do a little of each and a lot of none. Do the opposite.

Match page intent before you write a word

The other place teams burn budget is page format. Sam's test takes 30 seconds: type your target keyword into Google, look at the top three or four organic results, click them, and see what kind of page is ranking. That's the page intent. If "Calendly alternatives" returns ten listicles, your page is a listicle. If "best calendar scheduling tool" returns solution pages, yours is a solution page. Trying to rank a long essay against ten comparison tables is a category error.

Once you know the format, the rule is simple: blow the existing top result out of the water. If they wrote a top-10 list, you write a top-20. If they have screenshots, you have screenshots plus a live demo. If they buried the comparison, you lead with a feature table. Sam doesn't want to win narrowly — he wants the competition to look thin by comparison. That's how you survive a SERP where AI Overviews are siphoning everything above you.

The bigger trap on comparison pages is using the number-one slot to brag. Don't. Use it to show you understand the buyer's world. Position your product as number one, then explain what makes it different, who it's wrong for, what the trade-offs are, and which competitor wins in which scenario. Add the chart. Add the trial CTA. Add the customer video. That kind of page doesn't just rank — it converts the prospects AI Overviews can't help.

AI-written content can't carry a high-ACV deal

I asked Sam whether SEO still matters now that buyers are running 50 to 70 percent of their research through ChatGPT and Perplexity. His answer was the most honest one I've heard on the topic. SEO is a channel. Cold email is a channel. LinkedIn ads are a channel. AI changed all of them. The fundamentals didn't change.

What did change: the cost of cranking out generic content collapsed to roughly zero, which means everyone is doing it. Sam's example of what most AI-written B2B copy reads like: "20x your revenue with our all-in-one, all-singing, wizardry-dancing platform today. Supercharge your SaaS with our 360-degree CRM know-how." If you land on that page, you bounce in two seconds. Doesn't matter if it ranked. Didn't convert. Sam's read — and mine — is that prospects with real budget can smell robot copy from the headline.

The page that converts is the page that proves you understand the buyer's world. Their problems. Their objections. The exact language they use when they're frustrated. You get that from sales calls, customer success calls, Slack communities, Reddit threads — not from a prompt. SEO for B2B SaaS in the AI Overviews era is the same job it always was, just less forgiving of laziness. Good SEO and getting cited in AI answer engines increasingly run on the same fuel: real customer research, real points of view, real proof.

If you're leaning on AI to produce all that content for us, do we really think that's gonna convert? Even if you did rank a page, if someone lands on a solution page or a comparison page and it comes up with words like "supercharge your SaaS with our all-in-one wizardry-dancing platform," I'd bounce within two to three seconds. I could tell it's written by a robot. Most prospects with budget are gonna be the same. You can't fake understanding their world — their problems, their struggling moments, their objections.

Sam Dunning

What this means for your company

If you're a founder of a $5M to $20M B2B SaaS company, the next six months of organic search look like this. Top-of-funnel content keeps losing traffic to Google AI Overviews. Ad budgets get squeezed as funding tightens. The teams that win on organic in 2026 are the ones who already moved their SEO program to bottom of funnel seo — alternatives pages, solution pages by industry, jobs-to-be-done articles, comparison pages that don't read like a sales deck.

Three things to do this week. One: pull your last 12 months of organic traffic and tag every URL as top, middle, or bottom funnel. If more than a third of your SEO investment is top funnel, you're funding work that AI Overviews are deleting. Two: build the Google Sheet Sam described — offers, industries, competitors, jobs-to-be-done. Pick the five highest-intent keywords with even modest volume. Three: audit your top three commercial pages against the actual SERP. If the page intent doesn't match, rebuild the page before you write a new one.

None of this requires a new tool. It requires choosing pipeline over traffic, then committing the team to it.

Where the Pipeline Story Sprint comes in

If your SEO program is producing traffic that doesn't convert, the problem usually isn't your SEO team — it's that nobody at the top of the company has decided who you're for and what makes you different. Without that, every alternatives page sounds the same and every solution page reads like vendor mush.

The Pipeline Story Sprint is a 90-day, fixed-scope, fixed-price engagement for B2B SaaS founders between $5M and $20M. We sharpen positioning, rewrite the homepage, and build a content and SEO plan around buyers you can name. After the Sprint, your team can run the playbook — including the kind of bottom-of-funnel work Sam laid out — without me.

See if the Pipeline Story Sprint fits your team →

Listen to the full conversation
Unpacking Google's Leaks and AI Overviews: SEO Strategies with Sam Dunning

In this episode of Marketing Spark, Mark Evans interviews Sam Dunning, a content marketing and SEO expert, to dissect the recent leaks about Google's search algorithm and the rise of AI-generated content.

The conversation delves into the implications of these leaks, which reveal the ongoing importance of backlinks and brand relevance despite Google's previous statements.

Sam explains the critical impact of these changes on B2B SaaS companies and shares best practices for staying competitive in this volatile landscape.

They also discuss the future of SEO, emphasizing the shift towards bottom-of-the-funnel content and the diminishing value of top-funnel, AI-overviewed content.

The podcast highlights the importance of understanding high-intent prospects, leveraging detailed customer research, and creating content that genuinely resonates with target audiences.

As AI continues to transform SEO and content marketing, Sam offers practical advice on adapting strategies to maintain visibility and drive leads.

Thanks to this week’s sponsor, Landbot, AI Chatbot Generator for Conversational Experiences.