What It Takes to Scale a Brand in Agentic Search

What We Need to Grow, Episode 008: AI Visibility and What It Takes to Scale. A Founder Talk conversation with Steve Ward. Cleaned for readability; the words are the speakers' own.

FOUNDER TALK | 001: Creating for the Machine Layer A Conversation with Steve Ward of Haven by Design Stays

Caleb Pedosiuk: Steve, here we are in the wonderful world of digital pixel land. We met back in Kelowna, BC, through Dan Martell's elite coaching program, at the Social Selling Summit, getting to hear from his team and how they are doing their incredible work on the social front. It was really great to meet in person. I have loved getting to see the signal that you have been sending out online and just sharing your journey. My hope was that you would share a little bit more here about what you are working on and how that is going. Even since October 2025, so much has changed technologically, empowering more people to be able to build than ever before. So what have you been up to, what are you working on, and how is it going?

Steve Ward: I have been up to breaking everything. We hit a little bit of that parabolic push starting around October of last year. We were heavily trying to figure out how to leverage our time using Dan's buyback ladder, getting admin covered, then operations covered, and really bending our minds around how we grow this thing as fast as we possibly can and obliterate bottlenecks along the way. The goal was that when spring hit, like right now, we would have set ourselves up with the biggest foundation possible. Our business is very seasonal, so when May hits we can start seeing growth like I could only have imagined. I genuinely could not have imagined it back in January of last year.

It has been wild more recently, especially around the conversation about technology. I went pretty heavy into GPT initially. We used that for coming up with content on our backend, some SOPs, and we still use it sometimes for image generation, things like taking a picture of a unit and putting paint on the walls or swapping out the furniture digitally.

When I moved into Claude, it was really great because part of my background at my previous company was building a database in the FileMaker environment. So I had some understanding of how to structure a build, the difference between dev and prod, and some of those foundational concepts. Things that used to take forever translated into Claude very quickly. It has been a rapid adoption and honestly a lot of fun.

We now have a turnover app that all of our cleaners across four markets use for taking photos, documentation, and tracking stock levels. We also have an internal app that allows us to hand off between our VAs and EAs as they take different shifts throughout the day, so everyone knows what is on fire, what is high priority, and what is on the horizon. We keep the brain centralized. All of that came from rapid development. What would have taken me six to eight months in the past took two or three weeks.

Caleb: And you are in the short term rental space, correct? How many properties are you managing?

Steve: We have 23 currently and will have 24 by the end of May. Geographically, only 10 are near us, 10 more are about seven hours away, and one is in Sarasota, Florida. We are on the West Coast. So systems and communication matter enormously. Even before all the tech work, good systems with cleaners, clear standards, being very candid about expectations, and moving through cleaners rapidly have been a big part of how we manage so many properties across so many markets. And that is what started to break a little bit. We were hitting our threshold.

Caleb: Where is it breaking? Is it at the input level, where they are on different devices and things are not syncing?

Steve: We did run into some challenges because the cleaning staff all have their own preferred software. Usually it is Turno or some other platform built for cleaning companies. Asking them to adopt something new, we pushed it too far and got too complicated, which is easy to do in development. We stripped it back to simplicity. We had a level 10 meeting with Nicole and our executive assistant Hannah, and asked ourselves what the actual purpose of this app is. I was turning into IT support and an app developer. That is not the goal. The goal is to build something that creates efficiency and then move on to the next thing.

What is actually breaking for us right now is this. Hannah, our executive assistant, has been handling all communications. That allowed Nicole and me to stay laser focused on our family and on growing the business. Hannah is pregnant and looking to pull back, and I want to make sure everyone who works with us has the opportunity to live the life they want to live. For her right now, that means not being glued to her phone as she prepares for the baby.

We are backfilling what Hannah provided with VAs. We have one and we need two more. Nicole and I have been absorbing the weekend communications burden in the meantime, which has been a step backward. The counsel I have gotten from Dan, and honestly from the AI tools I use as a sounding board, is that it is not a step backward. It is a signal pointing us to where we need to focus next.

Caleb: You shared a video recently about being up all night working with Claude.

Steve: Claudenesia.

Caleb: Claudenesia, yes. The timing of that resonated with me. When you get into something and really start learning it, it pulls you in completely. We went through something similar. We rebuilt our site and redialed our focus over December into January, really retooling around what we are here to do. We put the dart in the center of the board, which for us is helping brands become the one that gets recommended. Because if you can be the brand that is recommended, it means you are doing all the foundational work really well underneath. Your signal is clear, you are consistent, the AI systems have consensus about who you are, and you are trustworthy in terms of where you are cited.

But when we started seeing how much was moving on the Claude front, we said, let us get in here and see how these tools work. And very quickly you realize how much you can build. That is where deep work matters. To be uninterrupted, especially when Claude Code is working on one task and another instance is working on something else. You are deep in it. Context switching can be useful, but doing it without distraction is important.

Steve: I had to slow myself down. I was doing too much. I would have two chats going, running two different concepts simultaneously while building, and I found myself ready for bed at seven every night because I was completely overtaxed. I had to reshape the way I approached development. One platform, one app, one solution brought to a completed point, and then stop. No more running multiple threads at once.

So I developed a habit. If an idea jumped into my brain while I was mid-build, I would record it, title it properly, capture the thinking, and then put it back down. Because I was burning through my attention. Switching contexts that many times over the course of an eight-hour session is brutal on the mind.

I need to ask you something, though, because it has been something I keep circling. I continually hear the conversation about SEO versus agentic search. I feel like what you just described may put you guys ahead of the curve. What are you seeing when it comes to the necessity of retooling for how people are searching via agentic search versus traditional SEO?

Caleb: Search engine optimization is about optimizing for a ranking. A lot of the time someone looks at their strategy and says, if we get enough backlinks, or backlinks from a really trustworthy site, and maybe we are paying for good PR to be featured in reputable publications, that lends credibility, which equates to trust, which pushes you higher in the rankings.

But if people are not scrolling search results anymore, if they are wanting answers, then you want to optimize for the answer, not just a high ranking score. Anyone who says SEO is dead just does not understand what is actually going on. I look at it more like this: a really strong SEO strategy becomes the foundation that gets activated and supercharged with intelligent AEO and GEO, answer engine optimization and generative engine optimization. A strong SEO foundation is something you lean on. But it does not automatically mean you are going to surface in an answer.

There are three things that matter most. They all start with C, which makes them easier to remember.

The first is clarity. Make sure you know clearly what it is you are talking about and that your brand communicates that cleanly. If a brand has gone through multiple iterations of product market fit, trying different messages and angles, and all of that still lives in the social content, blog posts, and articles connected to the brand, it sends a confusing signal. A human scrolling results is only going to hold a little information and will not go deep. But AI systems are constantly being updated and indexing everything. If you have multiple conflicting signals, the AI will have low confidence in recommending you. So the first priority is a really clear message.

The second is consistency. The signal has to be consistent across every surface the AI indexes, not inconsistent from one channel to the next.

The third is citability, or trustworthiness. That is where the SEO work plays in. A strong backlink strategy, content that is indexing on other authoritative platforms, all of that sends a much stronger signal to the machine layer.

At a foundational level, we call this a Signal Check, and you can run one at 79dev.com. You come in, share your website, your top competitors, and your most active social channel. Then we look at what is surfacing in AI search. If someone searches for a specific topic, are you showing up in that answer or is a competitor? What percentage of the time? Once you have that information you can reverse engineer the answers that are missing.

One of the biggest paradigm shifts for us over the last few months is this: we are primarily producing content for the machine layer, not the human layer. There are really great agencies producing content that follows trending topics to surf the algorithm. That is valuable work. There are others doing incredible, emotionally charged campaigns that make a lasting impression on humans. That is also valuable. That is just not what we do.

We produce content at the machine layer. If a human being never sees it, it does not matter. Because the brands we work with are solving meaningful problems in the world. Someone is going to ChatGPT or Claude and saying something like, hey, my dog has just been nuts lately. We are at the park and he is all over the place. My wife is frustrated because he wasn't doing this when he was younger. I think I need lessons. He is a Doberman. Is it too late? Are there any trainers in this area?

That prompt is 45 words long. It is specific, conversational, and layered. Compare that to the old behavior of typing "dog trainer near me" into a search bar. The old prompt produced a list. The new prompt produces an answer. And the brand that shows up inside that answer is the one that wins.

So if we are seeing people search for help with retroactive dog training, we create content for that brand that answers exactly that question in depth. We are not making a carousel slider for Instagram designed to interrupt someone's scroll. We are making content that lives quietly as an article or blog post and answers what people are actually looking for.

Once you have the foundational clarity and consistency in place, the creative tactics for content get interesting. It becomes about who you could be having collaborative conversations with. For example, we are doing some interesting work with architects right now. Architects and interior designers are a really strong ICP because they sit at the hub of decision making. They are between investors, builders, and end users. They are at the forefront of decisions that shape how people live. So for one client, we realized we should be inviting architects and interior designers into conversations, having them talk about building materials, tariffs, design philosophy, the future of living spaces. We structure those conversations in a way that answers the questions we know the models are being asked. Over time you build toward being the answer.

Steve: Specifically pinpointed. I would love to be a fly on the wall when you are working through this stuff. When you think about it, a lot of searches now are 40, 50, 60 words. Rambling, contextual questions rather than a tight Google keyword. I noticed this in myself. I get irritated trying to come up with a pinpoint Google search now. I want to just flow through the problem and get a real response. You pinpointed something that felt invisible to me until about five minutes ago.

Caleb: Yeah. I do not know if you use Superwhisper or Whisper Flow, but the moment someone realizes that all they have to do is push a button and just talk to their computer, there is no going back. I cannot even type anymore.

Steve: It is like being on the Enterprise. You just say, hey computer, and it makes you a cup of tea. I get genuinely frustrated now when I have to type. Something has shifted. The dopamine is registering differently.

Caleb: Typing is just a means to an end. Getting the intention from mind to tool. But the interesting thing is that most of the content that exists online was produced through that means. And that brings me to something that has been very much at the forefront of my mind.

There is that famous line attributed to Henry Ford: if I had asked people what they wanted, they would have said a faster horse. That paradigm has been sitting with me constantly while building with Claude. So many times I catch myself asking, am I trying to build a faster horse right now?

How many times do you see posts pop up saying, here is a prompt where you can create three months of content in 30 minutes? That is a faster horse. The reason it is a faster horse is because most of that content is still built on the assumption that you need to interrupt people in their lives, surface at the right moment, and hope it registers so they move closer to whatever funnel you have in mind.

But if you are solving a meaningful problem that someone would genuinely seek out, you want to surface at the answer point. That is the combustion engine. The paradigm shift.

There is an old idea about not putting new wine into old wineskins. If you pour unfermented wine into an old wineskin, the fermentation will expand it and it will burst. You lose both. That is what is happening with a lot of AI content strategy right now. People are using genuinely incredible new tools but pouring them into an old model. You see it in people's feeds. You think, great, now there is just more of what nobody actually wanted.

And here is the deeper thing. Most people do not actually want to be spending so much time on their phone. Most people need more fresh air, more time with the people they love, more time moving and resting and being at peace. What are we optimizing scrolling for? To learn things that, at the time you actually need them, are a single prompt away. If you were not burned out from over-indexing on passive consumption, you would be far more receptive to actually following through when the answer showed up.

Steve: That is very true. That is part of my own journey of stepping back. I keep asking myself the same question in different ways. Am I building a faster horse? It is like having a device in your hand that could have navigated something to the moon thirty years ago, and choosing to use it like a pen because that is what you already know how to do. That is a tough paradigm to break. Dan is always challenging us on this. He asks, you are having a great time in Claude, but is it actually making you money and buying back your time, or are you just enjoying yourself?

Caleb: So do you feel like coaching is on the horizon for you? You are a couple dozen units into this. For someone who is just getting started, a couple of units in and wondering what is possible, is that a direction you are moving toward?

Steve: I am already doing it. I am just not charging for it yet. Something in me resists being the STR guy, the Mr. Airbnb that everyone comes to for the same 10,000 questions. I would lose my mind. I like moving forward, not rehashing how to sign a lease with a landlord for the fiftieth time.

What I am genuinely drawn to is working with dads, mostly dads but parents generally, who feel like the career they are in is the only option but whose soul is telling them something different. I want to help them see what the path forward could look like. Honestly, it is not unlike what Dan does. I feel drawn in that direction and I am already doing it informally all the time.

Part of what drives this is that I was exactly that person. I am 40. I spent years feeling like I had missed it, like I should have started sooner. But I did not want to be wealthy for the money. I wanted to be with my family. I wanted bills to not be a source of stress. I wanted to be able to take people I love on good experiences. Very base level. And I want to tell people who are where I was that it is not too late.

That said, I am selective. I am not going to pour into someone who is not going to do the work, no matter how much they pay me. We have had more and more people reaching out, people in Idaho, people from the Elite group, a couple in Boise, a friend working on arbitrage in Tunisia. I still do not know if I want to be the STR guy, but I know I want to coach eventually.

Caleb: It seems like a lot of what you are already doing could be partially systematized. Even something like training a model on your own frameworks and understanding, a hybrid of your thinking and a tool that can help people get initial answers and guidance. We built two agents we use regularly. One is a voice agent that intelligently works through the questions we use in our discovery and onboarding process. After the call, the transcript gets handed to a synthesis agent, which uses a system prompt informed by the learnings and principles we keep refining. It takes the transcript, synthesizes it, maps it to a pre-designed HTML template, and produces a structured report. For our Clarity Report, it covers three categories: audience, brand, and content. It answers the five Ws for all of them. It works as a decision-forcing lens. If someone has answered a question vaguely or with low confidence, the report surfaces that and explains why.

The reason I bring this up is that instead of doing all of this manually, you could have someone call a number, have the voice agent ask the intake questions, take the transcript, synthesize it, and deliver tailored guidance back automatically. You would not have to be in the room for the initial intake at all.

Steve: That feels like a completely different level. Full disclosure, it is a little intimidating because I have not built my own agents yet. But it would save me an enormous amount of time. Is that the kind of thing you charge for? Is it a service? I would have no idea how to execute it.

Caleb: We are not currently going around building agent stacks for other businesses as a core service, but if it is something you wanted to build, we could work through it with you. Honestly, I think you are proficient enough now that a few parameters, maybe some screen recordings of how we did it, and a few hours in front of Claude Code would get you there.

Steve: I genuinely believe that. Four weeks ago Claude Code felt completely foreign to me. Now it does not. You just have to get past that initial resistance humans have toward unfamiliar things.

The question Dan keeps asking me is what do I need to say no to this week. That one sits with me. As Nicole and I have absorbed communications back, it has eaten into our deep work time. I will solve that. But then the question is, do I keep advising all these people informally? I am speaking at an Airbnb conference in two months on scaling, but I am showing up because I love scaling, not because it is Airbnb specifically. I keep asking people around me, when is it time to actually charge for the coaching? I love the model. I love what it can do for people. I love the margin.

Caleb: The margin, yeah.

Steve: It is just such a great combination. Genuine impact, real margin, and you actually enjoy every minute of it. I love sitting down with someone and just asking them what their goal is. They say a thing. You ask if it is actually doable. They say yes. You say great, now double the goal and cut the timeline in half. Their brain resets. They start asking the right questions. They finally move toward what they have always wanted. I have an absolute blast doing it. That is not Airbnb. That is something totally different.

Caleb: Maybe this is a bit theoretical, but I feel like AI is exceptional at execution. When a human being sees something they genuinely want to move toward and that spark is there, the execution layer is more powerful than it has ever been. But AI does not generate the spark. If there is no spark, all that capability is just potential sitting in the ether. How many times does someone half-heartedly ask an AI how to get in shape? The AI gives them a perfect plan. Nothing happens. The spark was not there.

But I think the spark can be lit through another person. That is one of the core reasons coaching works. When a coach says something and the person hears it and thinks, yeah, I could actually do that, something shifts. There is a kind of conception that happens. The circuit connects. The belief forms that it is possible. That is still a deeply human function. And maybe, increasingly, the how matters less. The how is a prompt away. The spark is the rare and valuable thing.

Steve: The how has never been more accessible. That framing really lands for me. It is a big part of why I want to focus on high-performing dads who are crushing their industry but do not believe they can do it on their own terms. I want to be the thing that flips that switch for them. That would be genuinely fulfilling work.

I am also very focused on sequencing. We hit the million dollar run rate last month. We did 87,000 in revenue. I want to keep that growing until we are solidly at a million a year. Then I will revisit where to direct my time and attention next. So coaching stays on the back burner for now. I will keep offering what I can informally, keep getting practice, and hopefully lay enough of a foundation that when the time comes to formalize it, the audience is already there. I have friends who call me and I tell them, that was a million dollar piece of advice you just got. They say they know. I say I will send you the bill. We both laugh.

Caleb: Why not actually put a page up? Put some rates on it. Even if it is just for new contacts coming into your world, when someone reaches out you can send them the link, let them see what you offer, and let them make a decision. It makes it real without requiring you to fully commit to a new service line. You do not have to have the conversation every time. The page has it.

And then maybe pick one day a week to go live. Not with a heavy agenda. Just something like, I am going live, come chat if you want. Talk about what you are working on. I have a few frameworks I wish I had known earlier. Come ask me anything. If no one shows up, that is fine too. You still produced something. You still sent the signal about what you are building and who it is for. And then you let the right people find you.

Steve: That makes a lot of sense. I do quarantine my limiting beliefs. I try to say them out loud because it helps to hear them. It has taken me over a year to go from 100 followers to 850. That is slow growth and I know it. I also know what I have to offer one on one. I know it is genuinely valuable. But it is hard to make that translate in a feed where cat videos win. I still feel like I am figuring out social. The live format resonates with me though. Even if one person comes on and says this is my goal, and I run them through a framework in real time, that could be a great conversation worth having publicly.

Caleb: That is exactly the format. Get on live. Have real conversations. You can go live across YouTube, LinkedIn, and Instagram simultaneously if you want. Invite people in to talk about what they are building, what they are figuring out. If the focus is tight and the invitation is clear, that creates a real surface area in a public way. That is what this kind of content does over time. It compounds.

Steve: More eyeballs, and the right ones. I will just say this: you are one of the few people who has been uniquely positioned on the sideline of what I have had to develop in myself over the last little while. There were some voice messages exchanged between the two of us during that period that were genuinely impactful for me. So thank you for being someone who shows up. If anyone ever asked what those first six months were like, I would say they were hard and I had to grow a lot. But I had some good people in my corner, and you are always one who comes to mind.

Caleb: Thanks, man. That means a lot. Alright, let us wrap up. Where can people find you online?

Steve: The best place is at scale.with.stephen on Instagram. S-T-E-P-H-E-N. That is where business, short term rentals, life, and family all come together. And if you are traveling to parts of Idaho or Sarasota, Florida, you can find our rentals at havenbydesignstays.com.

Caleb: Very cool. Thanks, Steve.