BRYAN MCANULTY [00:00:01]:

Hey, Chris. Welcome to the show.

CHRIS DUBOIS [00:00:03]:

Hey, thanks for having me, Bryan. I’m excited to be here.

BRYAN MCANULTY [00:00:06]:

Yeah, I’m glad to talk with you again. I just got to talk with you on your show, so I’m glad to get to know you better now on my show. Before Dynamic Agency OS existed, what frustrated you most about the industry, and how does your company solve that problem?

CHRIS DUBOIS [00:00:21]:

Before this, I was in the Army as an infantry officer. Then I shifted gears completely and entered the agency world, where I was a marketer for an agency. I became its head of operations, and then I became the CEO.

Through all of this time, I realized that, while we were usually good at marketing ourselves, we spent a lot of time focused on delivery and figuring out how to do really good work for our clients. That meant we had to rely heavily on referrals for new business.

When I started peeling back the curtain on that, I realized that because we were so focused on delivery, we weren’t looking at how we were different or how we could show up. That was why nobody knew to work with us until they actually talked with us.

Any marketing we were doing was just marketing. It was almost a series of random acts of marketing because we didn’t have a clear picture of who we were and why we were different that we could put out into the world to attract people. That is now the problem I solve for agencies.

BRYAN MCANULTY [00:01:38]:

Awesome. We’ve seen you say that positioning is not just about changing a headline; it is also about changing how the business actually operates. Can you explain that in more detail?

CHRIS DUBOIS [00:01:49]:

At the risk of throwing shade at some other positioning coaches—and I won’t use names—a lot of people view positioning exercises as simply changing the headline on your website.

That is part of positioning because positioning is something that lives completely in the buyer’s mind. We can’t control it, but we can influence it. One way we influence it is by putting our messaging out into the world. The challenge is that we can say all the best things, but it doesn’t matter if someone has no proof that this is how we actually operate.

Imagine that you want to be known as the fastest agency in the space and that you prioritize speed over everything else. The second someone starts working with you and a deliverable goes past its deadline—even if there was a perfectly reasonable cause—you are no longer seen as the fastest business. You lose credibility.

We need to decide how we are actually different and identify the structural choices we are willing to make that other people won’t. There are consequences to these choices. They might affect how we price, who we work with, or our delivery model. All of those things could be very different.

By making deliberate, intentional choices, we are saying no to certain people. We are not trying to be a generalist agency or attract everybody. We want to be the right fit for the right person at the right time.

Positioning helps us accomplish that. When we put out our message and change the headline, someone sees it and says, “This is 100% for me.” Your sales process becomes easier, you attract more of the right people, and you receive more referrals. There is a cascade of positive effects throughout the business.

BRYAN MCANULTY [00:03:41]:

It is like “show, don’t tell,” but it is more than that because you have to think about the consequences for your business. If you are going to say that this is who you are, how do you have to act to prove that it is true? What are the trade-offs and consequences?

That is a great point. If we return to the speed example and say that you have to deliver extremely quickly, then you need to deliver quickly even when it would otherwise be reasonable to take longer. That is how you earn the testimonial that says, “These people are so fast. I can’t believe it.” Then you continue to attract more people and further establish your positioning.

CHRIS DUBOIS [00:04:28]:

That is another thing a lot of agencies get wrong. They will put up a testimonial that says, “They were great, and I loved working with them.” That doesn’t really tell me anything about you. But if speed is your priority and every testimonial talks about speed, you become known as the agency that delivers that speed.

BRYAN MCANULTY [00:04:48]:

That is really great advice. It is also helpful to describe the referral problem you mentioned at the beginning. I think this is something a business has to constantly reevaluate.

I thought I was clear on the positioning of the different things we are doing with our products, but I only recently realized that we are doing things that are very different from our competitors. The differences are in the way we operate, but we weren’t telling people about them.

People come to us, find out, and are pleasantly surprised. They are really happy because we delivered this great service. But we are missing people who could discover us if we would simply tell them who we are in the first place.

CHRIS DUBOIS [00:05:35]:

One hundred percent. There are plenty of metaphors for this: you can’t read the label from inside the jar; Solomon’s paradox, named after the wise King Solomon, who couldn’t solve his own problems; or the cobbler’s children having no shoes. For whatever reason, it is difficult to see these things for ourselves.

Even in my own business, I very deliberately set things up to work with agencies earning less than $1 million. They are my primary target audience. Because of that, I have to charge less than I could charge multimillion-dollar businesses because these agencies don’t have huge budgets.

I have deliberately priced lower, but that means I have to structure my business so I can still make money while charging people less. There is a consequence to that decision.

I thought this was awesome and very deliberate. But when I start working with people, one of the things they point out more than anything else is that I will actually get my hands dirty and work on their marketing assets with them. They say, “I’ve never had a coach who actually does marketing work for me.”

I’m not doing their entire campaign. But if I know something will take me 30 minutes and might take them two or three hours because they have never done it before, that feels like a good teaching opportunity. I can help them see what good looks like, put the system in place, and help them move much faster and start bringing in business sooner.

Because so many clients have mentioned that, it now needs to become part of my positioning across my website, my content, and everything else I produce.

BRYAN MCANULTY [00:07:11]:

Real quick, Bryan here. You know me as the host of this podcast, but what you might not know is that I’m also the founder of Heights Platform. It is an all-in-one platform that more than 10,000 creators have used to build their online courses, communities, and digital product businesses.

We recently added some awesome updates to Heights AI to help you turn your idea into a viable business. Heights AI can build entire product offers, review your content, and even coach you on how to grow. You can try it free for 30 days. The link is in the description. Now, back to the podcast.

What are people telling you they love about your service that you aren’t correctly communicating publicly?

CHRIS DUBOIS [00:07:51]:

It can be difficult to find that. One exercise I like is to look five years down the road, before you even talk to your clients, and ask what they would say was the best outcome they received from working with you. What problem did you solve for them?

What do you think they would say? If you can answer that easily for yourself, go back to them and say, “We really think we were able to help you with speed. Would you be willing to give us a testimonial specifically about that and how it helped your business?”

You could even write it for them and invite them to change anything they think needs changing. If they are happy with it, you can use it. Then you get people saying the exact things you need, and it is still honest because you are only asking them to say something they are comfortable putting out there.

BRYAN MCANULTY [00:08:44]:

That is a great point. Since starting Dynamic Agency OS, can you think of one investment, big or small, that has had the biggest impact on your business?

CHRIS DUBOIS [00:08:52]:

That is a good question. I am a solopreneur. The investment I made in myself was probably building a robust AI chief of staff.

I can only do so much. Because of some of the pricing decisions I made, I need to be extremely efficient in my work. Otherwise, I can’t take on more clients, and I get stuck at a relatively low revenue ceiling.

By adding this chief of staff—I call mine a “chief of stuff” because it manages things rather than people—I can do much more with my business. I take everything I’m thinking and doing and remove the busywork. These are not necessarily menial tasks, but they include collecting client notes at the end of a call, ensuring everyone knows their action items, preparing my week, and preparing my podcast. These things need to happen, but they are busywork.

I can now do that work very quickly, so I spend more time with clients, thinking through their marketing. They receive more of my creative process and personal time as I help them build their accounts. It was a small monetary investment but a huge time investment to get it working exactly how I needed, with all the proper connectors and other components.

BRYAN MCANULTY [00:10:29]:

I want to talk more about AI now. If every business has access to the same magical AI agents and we can press a button that does almost any work for us, what makes your company different from somebody else’s?

I think we are already touching on the importance of processes and the way you work, but I would love to hear more of your perspective.

CHRIS DUBOIS [00:10:56]:

That is the million-dollar question. In my positioning model, there are 11 different levers we can pull. We should find three to five of those levers and stack them. That is what makes us truly different.

One lever is market focus: what industry do I want to work with? For all of these levers, the reasoning has to go beyond saying, “I want to work with these people because they have money,” “I enjoy working with them,” or “I’ve had clients like this before.” Why is it beneficial to them? There needs to be a reason for them to pick you.

I might use market focus as one lever. My delivery model could be another. Are we working asynchronously? Are we offering group coaching? Is it a done-for-you offer where I do everything? Those components of your delivery model matter to someone during the decision-making process.

I could also use my methodology—how I actually do the work—and add a strong risk reversal. What guarantee can we provide so people feel that more of the risk is on us than on them?

No matter how we use AI, if we use it to strengthen those levers and make them more pronounced, it becomes much easier to stand out instead of appearing the same as everyone else.

If asynchronous communication is part of our delivery model and something people in the market care about, I might build an AI bot that talks to my clients. It has my knowledge and can interact with them when I am not readily available. They always have that resource, so the service becomes completely asynchronous while they continue receiving attention.

The same idea could apply to risk reversal. If AI systems allow me to spend less time on each account, I can offer a stronger guarantee. I am not a fan of money-back guarantees, but if something doesn’t work out and I have to return the money or lose the account, I might have spent only five hours on the account rather than the 20 hours it would have taken without AI.

Instead of being 20 hours behind and needing another 20 hours to reach the same point with a new client, I can take on four more clients before reaching that point. That allows me to support a heavier guarantee if that is something customers care about.

There are many examples, but I would look at the levers you are already pulling to stand out and ask whether you can use AI to elevate a specific lever so that you stand out even more.

BRYAN MCANULTY [00:13:45]:

I think that is great advice. Everybody has to think bigger about how they can improve what they are offering, especially now that they can integrate AI in some way to help make it possible.

When did you first realize that AI could multiply your abilities? What clicked for you? Things are changing so quickly. Six months ago, you might have said it would be nice to have this chief of staff, but that it wouldn’t work, wasn’t reliable, or couldn’t be trusted. What changed, and when did you realize it could work for what you needed?

CHRIS DUBOIS [00:14:40]:

I think there were two different epiphanies. The first was probably around 2023. A lot of companies started letting people go because they thought one person with AI could do the work of five. Even then, that was probably a stretch, but the idea was that AI was a force multiplier. If it could do that much more work, companies thought they could save a lot of money by reducing their team size and having one person do the work.

When I saw so many companies doing that, my immediate response was, “Why?” It didn’t make sense to me. If AI is a multiplier, why wouldn’t I give five people that five-times multiplier and get 25 times the effort inside our company? We can do so much more for the same price.

I might not grow the team as quickly, but I don’t need to let people go. The company that keeps only one team member is now merely as efficient as every other company. You are saving money, but what are you doing with it?

That was one thing that got me thinking differently about AI. How do we use it to multiply our effectiveness and efficiency?

BRYAN MCANULTY [00:15:50]:

I completely agree. It is not only about saving a little money. How can we do better than ever before now that everybody can use this capability?

CHRIS DUBOIS [00:16:03]:

The second epiphany was realizing how important context is for making AI work. The people I heard complaining most about hallucinations and poor results were trying to accomplish everything with a single prompt. They weren’t using a system.

My chief of staff has full access to everything I touch. Every meeting I attend has a Fathom recording, but I also have Granola running while I add notes. If it is a really important meeting, I take handwritten notes because, for some reason, I still do that during meetings. I will dictate those notes into a Google document, which also gets added. The system has access to everything I’m thinking that wasn’t verbally stated during the call.

I don’t think I have ever had a hallucination while using my chief of staff. I give it very specific things to answer based on work I am already doing. It is trained on my processes and systems, which I know can produce a manual result. I’m not trying to get it to create something new out of thin air.

Using that type of system is like night and day compared with people who enter one prompt and receive terrible results.

BRYAN MCANULTY [00:17:24]:

If five employees can each become five times more productive with AI, how does a founder or entrepreneur need to rethink roles and responsibilities?

CHRIS DUBOIS [00:17:37]:

One quick point first: five times is probably a fair estimate, but when everybody works on it together and shares their results, it becomes six or seven times better. We share our findings. It is not one person learning how to do something; it is one person getting a great result, showing everybody else, and saving them the time required to learn it independently.

BRYAN MCANULTY [00:18:04]:

Even if it is nothing more than saying, “Hey, everyone, look at this,” it makes everybody aware that something is now possible.

CHRIS DUBOIS [00:18:12]:

Right. I don’t code at all. I have no coding experience. One of my friends, who worked at the agency I was running, approaches everything with a developer-first mentality. Sometimes he will use a term that I can then explore.

For example, I didn’t know what a cron job was. I was asking, “How do I get this thing to fire at certain times?” He said, “It’s a cron job.” Hearing that term was all I needed to unlock how to set things up so they fired at the correct times. Your point is 100% right.

As for how founders should rethink the structure of the work and its positions, instead of hiring entire teams to do everything, I would consider making one person the head of a specific topic.

Even before AI became prominent, I played with the idea that companies always have department heads. You might have a head of marketing with many marketers working beneath that person. But you rarely have a head of copywriting who runs how all the copywriting is done at the company.

That person doesn’t have to see every piece of copy, but if someone has a question, they are the source to consult. You can create two paths: a specialist path and a leadership path.

The specialist path works really well with AI. I can become extremely good at a specific marketing skill, build all the agents and systems related to it, and manage an AI team. I still get some leadership experience, and I can give those resources to other team members. We are building an AI team beneath the humans, who can manage it because they are already subject-matter experts.

BRYAN MCANULTY [00:20:11]:

There is a lot here. Someone listening might say, “I sent some prompts to ChatGPT or Claude, and it gives me a message back, but what you are talking about sounds very different.”

There is a lot to understand to bridge that gap and use these tools in this different way of working. What about solo entrepreneurs or people with a virtual assistant or a couple of people helping them occasionally? What happens when they start moving faster than the people they work with?

Do you have any recommendations for helping a team or employees understand how to move beyond doing tasks directly and effectively direct AI to help them?

CHRIS DUBOIS [00:21:12]:

I think you need some type of learning system for your AI. One simple thing someone coached me on was to consider how we would normally bring a project to life. If an agency were hired to create something from nothing, what would happen first?

We would create a PRD, or product requirements document. That is our planning process. I talk with the AI and say, “This is everything I want to build. Before you build anything, let’s discuss what it could look like.”

We go back and forth and have a conversation. The final output is a massive document containing everything that needs to happen during the build process.

If that document matches the outcomes we want and contains all our processes, we can feed it our documentation explaining how our company performs the work. Maybe you had a call with the CEO who gave you instructions to make something in a very specific way. You can provide that call recording so the AI has context about why the project exists.

Once you have the PRD, you build the project, run tests, and keep making adjustments until you receive the exact result you want. You establish a gold standard for what the final output should look like. The AI is trained to use that gold standard to evaluate everything else it produces, and then you are off to the races.

It takes a project-management mentality. You can’t simply say, “Build this.” You need to ask where to start, make sure the requirements are set, build it, test it, validate it, and then move forward.

BRYAN MCANULTY [00:22:49]:

I think that is probably the biggest thing people overlook. We are presented with a prompt window and a text box, so someone says, “I need this thing,” presses send, and receives something they didn’t want.

But did you explain what you wanted? You have to spend time communicating everything that matters and what the actual result has to look like.

That could sound tedious, but if you were working with an efficient team, you would be doing the same thing with them. If you handed the task to somebody without doing that, you would spend all day going back and forth on calls. You are essentially sharpening the axe.

The benefit of AI is that if you aren’t sure what to communicate, you can ask it. Ask, “Is this clear to you? Do you have any questions? Ask me more questions to clarify what I want.” Once you have done that and everything is set, you can tell it to build, and you will get what you want.

CHRIS DUBOIS [00:24:02]:

In Claude Code specifically, you can use “ultrathink.” I don’t know what other tools use for this, but if you type the word “ultrathink,” it turns into a rainbow of colors and gets the system to think much more deeply about the topic.

You can say, “Use ultrathink, enter planning mode, and be comprehensive. Ask me every possible question you can think of to make sure we have the exact plan we need.” You can get some very good results.

As a society, we have spent so much time typing that we have trained ourselves to self-edit as we type. We want to make our messages as concise as possible, so we think about the fewest words required to communicate something.

I recommend using a tool like Wispr Flow or another dictation tool and simply spilling every thought onto the page. Get everything out. When you dictate, you say everything on your mind. That works to your benefit because the AI receives much more context about what you are thinking.

You aren’t self-editing. Most of us aren’t great at public speaking anyway, so getting all the words onto the page helps substantially in producing a better output.

BRYAN MCANULTY [00:25:22]:

I completely agree. I do the same thing. I bought a small clip-on microphone so I can walk away from the computer and talk for a while. Then I give that entire mess to the AI, and it works better.

You are right that it contains more than I would provide if I sat there editing and thinking about how to say something perfectly so it would be understood. You can ramble, correct yourself, and eventually provide all the details.

Unlike a human, the AI doesn’t have to sit there listening to you talk for half an hour or an hour. You have to make things clear for a person, but AI can process the text and figure it out.

CHRIS DUBOIS [00:26:07]:

My content process starts with SEO research and reviewing voice-of-the-customer material from places like Reddit. I find the topics I want to discuss. Promoting my positioning is obviously one critical component, but I also identify those other topics.

I have an SEO research tool that works through an API, and Claude Code has access to it. It checks how difficult a keyword is to rank for, the search volume, and other relevant information.

I then review the titles and the angle we want to take. The system is already trained on how I think about these subjects. It finds three to five questions for each topic that would allow it to create a unique article if it had my answers as a subject-matter expert.

I usually work in batches of four to six articles. The system puts every article and its three to six questions into an interview document in Google Docs. I open that document on my phone, go for a walk, and dictate my answers to each question.

When I return, it writes all my content, and then I edit it. The edits continue improving because I always feed the articles back into the system and say, “Here is what I changed. Log and store that in your memory.”

Because it has my actual thoughts, I’m not worried about someone complaining that it is AI content. Yes, I used AI to help write it, but they are all thoughts I personally had and wanted to put into the world. This is the fastest way for me to provide content that could help your business. I am comfortable with that, even if someone doesn’t like an em dash here and there.

BRYAN MCANULTY [00:27:52]:

I agree. I have been following essentially the same process. That is the difference between saying, “Write me an article,” which will produce absolute slop, and saying, “This is what we need to research, and this is what we are trying to accomplish. Now ask me a series of questions.”

I did this for one of our products two days ago. The system came up with 58 questions. I turned on dictation, went through the questions, and answered all of them. Now it knows what I think and why I would approach the subject in a particular way.

It is all coming from me. The AI is simply transforming what I said into a better blog-post format.

CHRIS DUBOIS [00:28:46]:

One hundred percent. When I think about how I used to create content, this is so much better. I wish we’d had these tools before. At the same time, I wouldn’t understand what makes a good article if I hadn’t first gone through the process manually.

BRYAN MCANULTY [00:29:02]:

That is a very important caveat. The areas where you can move the fastest are the areas where you already have the most knowledge and understanding.

If you don’t have that knowledge, you can make bad things quickly. You still have to learn and determine what makes a good article, what people care about or want to read, and what they are searching for.

Once you have a clear and deep understanding, you can move extremely quickly with these tools. Everybody has something they are particularly skilled at, and those are the areas where they can move fastest.

CHRIS DUBOIS [00:29:44]:

One hundred percent. We are in a different world now, but it is awesome. There is so much we can do. I would not have been able to accomplish nearly as much with my solo business without these resources.

BRYAN MCANULTY [00:30:12]:

Absolutely. It is an incredibly exciting time for entrepreneurs.

I want to move into a quick lightning round. I’m looking for a 10-second answer or a quick take for each question. Is replacing employees the worst way to use AI?

CHRIS DUBOIS [00:30:27]:

I would say so.

BRYAN MCANULTY [00:30:34]:

Should every employee be expected to use AI in 2026?

CHRIS DUBOIS [00:30:41]:

Yes, unless you are doing a very manual job that AI simply can’t replicate yet. You probably need to be using it in some way. Otherwise, you are leaving money on the table.

BRYAN MCANULTY [00:30:52]:

If you do any kind of work with a computer, you should be using it. What is one task humans should never delegate to AI?

CHRIS DUBOIS [00:30:56]:

Sending a birthday note to your mom.

BRYAN MCANULTY [00:31:06]:

That is a good one. Will small, AI-powered teams outperform large enterprises?

CHRIS DUBOIS [00:31:16]:

I think so. Consider how hard it is for enterprise companies to make any decision because of all the components that must move and align and all the red tape they must navigate. A small team can get a lot done now, and with that multiplier effect, it can crush them.

BRYAN MCANULTY [00:31:34]:

Is AI making most agencies interchangeable?

CHRIS DUBOIS [00:31:40]:

Yes, especially the ones that are not using it correctly to reinforce their positioning, make themselves stand out, and do the other things we discussed earlier.

AI will always default to the average. Unless you are doing something unique with AI, it will not help you stand out.

BRYAN MCANULTY [00:31:59]:

What is the biggest lie people believe about positioning?

CHRIS DUBOIS [00:32:03]:

That changing the headline on their website is enough.

BRYAN MCANULTY [00:32:07]:

On this show, I like to have every guest ask a question of the audience. If you could ask our audience anything—whether it is something you are curious about or something you want to get people thinking about—what would it be?

CHRIS DUBOIS [00:32:20]:

What intentional decision have you made about how you want to do business that is actually unique, and have you been using it to help yourself stand out?

That is probably a run-on question, but find your uniqueness and put it out there. It starts by asking what makes you different. Hopefully, it is something intentional that you have done with your business.

BRYAN MCANULTY [00:32:46]:

I think that is a really good thing to consider. Chris, it was great talking with you. Before we go, where else can people find you online?

CHRIS DUBOIS [00:32:54]:

You can learn more about me at Dynamic Agency OS. If you simply want to chat, LinkedIn is the place to go.

BRYAN MCANULTY [00:33:02]:

Awesome. Thanks, Chris.

CHRIS DUBOIS [00:33:03]:

Thanks for having me.

BRYAN MCANULTY [00:33:05]:

I’d like to take a moment to invite you to join our free community of more than 5,000 creators at CreatorClimb.com.

If you enjoyed this episode and want to hear more, check out the Heights Platform YouTube channel every Tuesday at 9:00 a.m. U.S. Central Time. To get notified when new episodes are released, join our newsletter at TheCreatorsAdventure.com.

Until then, keep learning, and I’ll see you in the next episode.