Stephan Bajaio has been in search for two decades, helped put some of the first e-commerce sites on the internet, co-founded Conductor and built its professional services team to 65 global SEOs, did a stint as a CMO to see how the other half lives, and now runs Vibe Logic with his co-founder Trevor Stolber. He came on The Unscripted SEO Interview and spent an hour arguing that the industry’s biggest problem is not Google, not AI, and not executives. It is us.
“It’s our fault”
Jeremy opened with the usual joke about SEO being declared dead for the seventh time. Stephan corrected him to seventeen — he wrote a Halloween piece stacking up the zombies of every SEO obituary ever published — and then refused to leave it as a joke.
“The reality is that we’ve done a piss poor job of defining ourselves within organizations. We’ve let others define us for us. We still are in an argument right now about what we are, a GEO, AIO — like that’s our fault. That’s no one else’s fault.” — Stephan Bajaio
He was Conductor’s chief evangelist, which he says meant evangelizing search itself, not just the product. His verdict on how that went for the industry is unsparing: SEOs are some of the smartest people in the rooms they are in, and they act like idiots when it comes to defining their own field. The GEO-versus-AIO naming fight is not a branding curiosity. It is a symptom.
So what is the job, actually
Asked to define it, Stephan strips it down. The optimization part is what we do — take data, take content, and find opportunities to make whatever ingests and judges that content understand it better. Everything else is complication: algorithms, LLMs, voice search, universal search, the fact that GA4 broke your reporting.
“Your job is to look at the wisdom of this organization, the audience that needs that wisdom, and finding ways to connect those two. Search engine is just the medium by which it’s done.” — Stephan Bajaio
The failure, in his telling, is that the channel held an amazing data set — genuine supply-and-demand intelligence about what users want and what actually ranks — and the industry treated it as a channel instead of as marketing information. That turned every insight into a hammer looking for a nail: I need to rank for that, I need to rank for that. Which sets a trap, because if you do not rank, you failed — and that is not how LLMs work, and increasingly not how search engines work either. They look at whether you are cited and referenced.
The lonely island
This is the part any in-house SEO will feel in their teeth. Stephan describes every SEO as stranded on a one-palm-tree island; when they spot another SEO they think they are saved, and then discover that person is on their own island too. It does not improve at enterprise scale — the number of SEOs at a big company is myopic against what they are being asked to accomplish.
The structural problem is that the SEO usually does not create the content and does not do the dev. If the writer, the dev team, or the executive setting brand direction is not on board, you are finished.
“It doesn’t actually matter how smart you are, it matters how capable your organization is to actually accomplish the smarts you have.” — Stephan Bajaio
You could have Larry Page call your cell phone and whisper the algorithm to you and still fail 100% of the time if those people do not move. Jeremy recognized the trap from in-house years at Raven Tools and TapClicks — building complete, spec’d-out product marketing dev tasks in Jira exactly as asked, then watching the resources get allocated elsewhere in the meeting, every time.
Conflicting KPIs, and the fix
Stephan’s diagnosis is that marketing departments organized by channel produce KPIs that conflict by nature. Take the simplest case. Your content writer is measured on volume — five blog posts a month. You come back and say pages six through nine need optimization, at least another paragraph each. Have you helped them hit their number? No. You have actively distracted them from the thing that gets them the pat on the back, the bonus, or just the sense of keeping up. And it is not the CMO’s job to arbitrate that, even though it should be, because the CMO is focused on vision and direction rather than operations.
The practical move he has landed on after twenty years:
Do not add items six through nine to someone’s to-do list. Find out what is already number one through four on that list, and make those better. Go to the article they are already writing, offer three things that will make it perform, and then — critically — show them the performance so they can take the credit with their own boss. Now you are pulling the same direction instead of competing. He is honest that this is a luxury to describe and much harder to do, and says that difficulty is exactly why the industry is not where it should be, why rates are low, and why people get away with declaring it dead every time something new shows up.
Speaking exec, not speaking tech
Part of the problem is linguistic. Search arrived, companies looked around for whoever knew HTML, and that person became the SEO. But speaking tech and speaking exec are two different languages. Stephan — who says he is far more fluent in exec than tech, and hands the technical side to Trevor, the “logic” in Vibe Logic — puts it bluntly: say “canonical” in an executive meeting and you will raise more wristwatches than eyebrows.
Executives do not care about the technical layer, and here is the uncomfortable part: they do not care about the channel either. When Stephan became a CMO he assumed SEO would be his channel. It was not. Someone was breathing down his neck for leads and he did not care where they came from. That is why he talks in supply and demand — the linguistics of business, not the linguistics of canonical tags.
Algorithm updates, and going deeper instead
Jeremy floated a quote from Chris Tweeden of Space Bar Collective: talented SEOs look forward to Google algorithm updates. Stephan says it depends on which kind of SEO you are. Trevor gets dopamine from picking apart a floating math equation in the sky. Stephan does not. He looks forward to updates only when he knows the work has been done right — and he would dread one if he knew a client was publishing all-AI content, because that becomes an “I told you so” conversation he never wanted to have.
What he focuses on instead is depth. He wears a t-shirt that says if you’re not helping people, you’re just selling stuff, and treats it as a marketing mantra: give away enough and you earn trust — from the engines, from the LLMs, and from the audience. His example is the sharpest thing in the episode.
“Someone’s bullet point is someone else’s entire universe.” — Stephan Bajaio
He was working with a healthcare company whose page on qualifying life events listed divorce as a bullet point. One bullet. He wrote an entire copy brief on divorce, because people who are getting divorced have real questions: Can I keep my spouse on my insurance? For how long? Do I have to tell the insurance company? What if we are only separated? What if I do not want my spouse on it? Does answering all of that convert every reader into a customer? No. Does it associate the brand with having given the answer, somewhere in someone’s head, in a way that finds its way back later? That is the bet worth making — and it survives whatever the SERP does next, because whatever comes next still has to interpret something. The question is what you are giving it to interpret.
Interviews are the best keyword research
Jeremy’s contribution to the actionable half: stop starting with the tool. He scheduled an interview with the guy who physically installs precast concrete walls and got the real questions — do we have space for this, do we need a cleared pad, and measure three times because concrete does not change dimension after it is poured.
“Get two experts riffing on a subject, they will bring up the most cogent pain points. They will address the elephants in the room.” — Jeremy Rivera
He also copped to the opposite: an agency job where he produced dozens of articles for a client he never spoke to, alongside a content team who also never spoke to them, with all interaction routed through a customer support rep he never spoke to either. Everyone in a silo, and then we wonder why we are disconnected from clients.
Stephan’s version, turned into a process
He took that and made it operational on the spot, and it is worth following step by step:
- Before the interview, ask an LLM what the most frequent questions are that you should ask an installer of concrete walls. Now you have a base.
- Check Semrush or Ahrefs for additional frequently-asked questions — maybe not for the installer, maybe for the salesperson.
- Record the conversation and run it through whatever transcription you already have. An entire transcription industry got wiped out by a feature; use it.
- Mine the transcript for pain points, considerations, and prerequisites. Ask whether they track how often a crew has to return to a site because it was not ready — if that metric exists, get your hands on it, because it is your ROI story.
- Turn the prerequisites into an asset: a readiness checklist or calculator. If you cannot write a line of code, throw it to Claude and then QA the hell out of it.
- Ship it as something customer service sends before the installer arrives. Now it is an unwritten written contract — they checked the boxes, and if they say they did and did not, the blame lands where it belongs.
- Do not stop there. Attach the question bubbles that come up around each step to the content you just created — FAQs pulled directly from the interview, living on the site where engines and customers can both find them.
Will the checklist rank? Probably not. Stephan does not care, and that is the point:
“The intention wasn’t just to build content so it could be found through the model by which you want it to be found.” The flawed logic in SEO was believing the search engine was the only source, and that anything not arriving through it has no value. He would happily build a page that gets zero organic visits if it becomes the number-one newsletter piece or a modal every prospect sees on the homepage. Reporting only organic traffic on a piece of content — no direct, no referral, no other channels — and then claiming a fraction of the credit because it has to fall cleanly in your channel, he calls insane. It did not exist until you told them to build it.
Authorship is coming back
On where this goes next, Stephan thinks the real idea of authorship shows up again. Google Plus had authorship right and then had it wrong, and he suspects the author profile concept still lives somewhere in the back end — they know what you have produced. He points at John Mueller’s remark that if you have not produced twenty pieces of content about something you are not an expert on it, with the usual caveat about how offhand comments become canon.
His read: you are going to have to show up as a human, with a human name behind the work. The team account does not count, and the anonymous I posted it therefore it is, and my brand carries the value era is over. He is considerably less impressed with the schema property for marking content as AI-written — he reads it as a new disavow, asking practitioners to beat the bushes so the engine can identify signals it wants. Jeremy’s counteroffer: remove “not provided” and I’ll do it. Either way, the practical instruction is unglamorous and concrete — go back and fix your author profiles, your about pages, the internal linking between them, and the entities that establish who you are.
The funnel is an hourglass
He also thinks the funnel is the wrong shape, and that marketers as a whole shot themselves in the foot with it. It is an hourglass, and everything ends up at the bottom again — unless you are genuinely in the business of selling something to someone once. Coming from SaaS and subscription, the post-purchase experience is what perpetuates back into referral.
Most companies stop at “got you.” Samsung and Apple do not — they invest in the post-purchase questions, because they know people search “what’s that blue thing” and that misinformation will land in an answer box and damage the brand. Everyone else leaves it. And now those messy, non-keyword-shaped questions are arriving as LLM prompts, which do not come in clean, measurable monthly-search-volume packaging. Which is how you end up with an executive asking why these guys are talking about our pricing, or why that bad review keeps surfacing. His prescription: go back to your chat logs, find the most asked questions and the most plaguing problems — including the ones you already solved. Conductor could not do global rankings once; prospects were still repeating that a year after it shipped. Wherever someone last touched your brand is where you left them.
Web presence intelligence
Vibe Logic came out of Stephan meeting Trevor Stolber during COVID through a Hook Agency meetup organized by Tim Brown. Trevor is the technical half — high school at twelve, two points shy of Mensa, seventeen pending patents, in search on the side for 26 years. Stephan is, by his own description, the vibe.
The framework they built is what he now calls web presence intelligence — renamed from “search presence intelligence” because search was too narrow. They look at the supply side: LLMs and search engines are going to show a set of results to your audience. Is your response to treat that as a ranking opportunity, or to recognize it as a multi-channel influence opportunity? He describes the output as a roulette board — helping a client decide where to place bets by probability of getting in front of the audience.
“Be where decisions are formed, not made.” — Stephan Bajaio
You form an opinion on Reddit. You form one reading an article. You form one on a partner site. Because marketing is siloed by KPI, the display person will not see that opportunity, the social person will not see it, and neither will demand gen or performance. They see vertically; Vibe Logic looks at it horizontally, the way the consumer actually moves.
Diagnosis before prescription
One more thing worth stealing: how they open an engagement. They do not walk in and tell a company what it needs. They run a diagnosis in which the company diagnoses itself — and what surfaces is usually dissonance. The CMO believes one thing, the head of product believes another, and that conversation was never going to happen naturally inside the company. Facilitating it is not SEO. The next meeting delivers a prioritized roadmap with an explicit “if you want to do this without us, go.” Home remedy or ongoing treatment, the client’s call.
The reason is not modesty. If a client arrives at their own conclusion about why you are needed, the relationship stops being adversarial — and you avoid the failure Stephan has lived through: delivering copy brief after copy brief into a company with exactly one writer already tied to seven hundred other things. You fulfilled the contract. The needle did not move, because you gave them work that did not map to their capability. Vendorships are a dime a dozen; the goal is partnership.
The one thing he hires for
Ask Stephan what trait matters most and he does not say technical skill. Anyone can be taught aspects of search. He hires for passion — and points at Jeremy going out to interview a concrete installer, not because it paid more but because it was what he felt should be done, as the tell. He also notes that passion does not look the same on everyone, so learn to recognize how it shows up in the people around you rather than squashing it.
Algorithms will change. LLMs will blow up your work. Executives will chase new shinies, devs will make you validate everything ten ways, and content writers will tell you that you are trampling their brand voice. Passion is what carries you through, which is why he is twenty years in and doubling down by building an agency in an industry everyone keeps burying — not an SEO agency, he insists, but a visibility agency covering search, LLMs, and any surface where people look for solutions.
Connect with Stephan
Find Stephan at vibelogic.com, and on LinkedIn as Stephan Bajaio — his self-described drug of choice, and where he is happy to spitball on a business problem, a search problem, or a communications problem where you cannot get the message across to an executive who does not get it.
Listen to the full episode on Castos. For more conversations like this, explore the Unscripted SEO topical hubs or try the keyword-research tool at SEO Arcade.
