As AI becomes part of nearly every business discipline, choosing the right partner is getting more complicated. The real question for Kansas City businesses isn’t who calls themselves an AI agency, but who understands what AI is actually changing about the work.
Lance Hemenway
“AI services” may be one of the least useful descriptions of a business service I’ve encountered in a while. It can mean building software, automating internal processes, improving marketing, analyzing data, changing customer service, developing products or helping employees work differently. Those aren’t variations of the same assignment. In some cases, they barely belong in the same industry.
The agency business has been through versions of this before. There was a period when companies went looking for internet agencies, then digital agencies, then social media agencies. Eventually the technology stopped being the category and became part of how nearly every competent agency worked. I suspect AI is headed in the same direction, only much faster. That’s why I’m not convinced most businesses should begin by searching for an “AI agency.” The more useful question is which part of the business AI is actually changing and whether the people you’re hiring understand that part of the business well enough to improve it.
If I were looking for AI services in Kansas City today, I’d start by getting rid of the word AI for a moment and writing down what I actually want to change. Maybe the sales team spends too much time researching prospects. Customer service is answering the same questions hundreds of times. Employees can’t find information scattered across internal systems. Marketing production is too slow. The company wants to add intelligence to a product or rethink how a particular process works.
Those are much better starting points because they expose what kind of expertise is really required. I’ve watched enough technology waves move through marketing to be wary when the technology itself becomes the strategy. Businesses went through periods when they needed a website because everyone needed a website, social media because everyone needed social media and marketing automation because everyone needed marketing automation. Eventually the conversation matured and became less about owning the technology and more about what the company was trying to accomplish with it.
We’re beginning to see the same shift with AI. A 2026 PwC study of more than 1,200 senior executives found that companies realizing the greatest economic gains from AI were twice as likely to redesign workflows around it rather than simply add AI tools to existing processes. I think that’s an important distinction. Giving everybody another application isn’t necessarily transformation. Sometimes the more valuable question is whether the underlying work should still be done the same way at all.
One thing I wouldn’t assume is that AI expertise automatically makes someone an expert in the business problem they’re applying it to. I’ve seen versions of that mistake throughout my career in digital marketing. Knowing the technology and knowing what a customer will respond to are two different things, just as knowing how to automate a process and knowing whether that process makes sense in the first place are two different things. AI doesn’t make those distinctions disappear.
If the problem is internal operations, the important expertise may be workflow design, systems integration, data and security. OpenAI’s enterprise research has documented organizations moving AI into repeatable, multi-step workflows rather than limiting it to isolated tasks. Once AI starts touching real business processes, somebody needs to understand where information lives, which systems communicate, what employees should approve and what happens when the technology gets something wrong. If the assignment is customer-facing, the necessary experience changes again.
That’s why I wouldn’t get overly attached to the “AI agency” label. These tools can accelerate an extraordinary amount of work, but speed isn’t particularly helpful when the underlying assumptions are wrong. A company can automate a bad process, produce mediocre marketing faster or build technology around something customers didn’t particularly want in the first place.
I’d also want to hear an agency tell me that something isn’t worth doing. There’s an enormous amount of AI technology available right now, and that makes it surprisingly easy to spend money solving problems a business doesn’t actually have. An agency that can explain why it wouldn’t use AI in a particular situation may tell me more about its judgment than a presentation showing me 25 things it can build.
I remember when putting “digital” in front of an agency’s name actually told you something. It distinguished that company from traditional agencies that were still primarily built around television, radio, print and outdoor. Eventually the distinction became almost meaningless because digital became part of practically everything an agency did.
I think AI is likely to follow a similar path. For the next few years, companies will advertise AI services because businesses are actively looking for them. Eventually, though, using AI will probably become an expected capability rather than a specialization. Web developers will use it while building websites. Media agencies will use it to analyze campaigns and audiences. CRM consultants will use it in automation and customer data. PR firms will use it for research and monitoring. Software developers will build it directly into products, while marketing agencies will have to understand not only how to use AI internally but how it is changing customer discovery and decision-making.
At that point, the interesting question won’t be whether an agency uses AI. It will be whether the agency understands what AI has changed about the work it’s being hired to do. A marketing agency can use AI to produce content faster without understanding how AI is changing the way customers research companies. A development company can use AI to write code faster without necessarily knowing how a client’s workflow should be redesigned. A consultancy can put together an AI strategy without having the operational experience to implement it. Using AI and understanding what AI changes aren’t quite the same thing.
Learn more: What’s the Difference Between an Agency That Uses AI Tools and One That Builds AI Visibility?
Kansas City has always had an interesting mix of agencies, technology companies, software developers and consultancies, and AI is beginning to blur some of the lines between them. I’d be much more interested in whether the people I’m talking to have experience with the business problem I’m trying to solve and can explain where AI genuinely changes their approach.
I’d also pay attention to the questions they ask. If most of the conversation is about models, platforms and tools before anyone has spent much time understanding the business, I’d probably be cautious. The technology matters, but I want to know that somebody understands the customer, workflow, system or business outcome before deciding what should be built around it. Sometimes the smartest technology decision may be a fairly simple one, and I’d rather work with people comfortable saying that than people who need every engagement to become an AI transformation project.
I suspect “AI agency” will eventually become about as descriptive as “digital agency.” It tells you something today, but probably not enough to make a hiring decision, and perhaps even less a few years from now. The category will mature, specialties will become clearer and AI will become embedded deeply enough in existing disciplines that separating “AI work” from everything else may start to feel artificial.
If I were hiring for AI services in Kansas City today, I’d begin with the part of my business that’s changing and find people who understand that problem exceptionally well. Then I’d want to know whether they understand AI deeply enough to approach that problem differently than they could have a few years ago. That may lead to a software developer, technology consultancy, automation specialist, marketing agency or some combination of them. As AI becomes part of more of the work businesses already do, I suspect that combination of judgment, domain experience and technical capability will matter a lot more than whatever agencies decide to call themselves.
Lance Hemenway is CEO & Founder at Fireball Agency, a Kansas City-based marketing agency helping businesses build AI authority and visibility. Connect with Lance on LinkedIn.
Over the past two decades, Lance has helped launch, grow, and scale digital businesses during some of the internet’s most transformative moments.
