I’m going to tell you something that might get me kicked out of Silicon Valley. You don’t need to be a machine learning PhD to lead an AI transformation. In fact, sometimes it’s better if you’re not.
I’ve seen it happen a dozen times. A company gets a bad case of AI fever. The board is screaming about it, the competition is launching “AI-powered” everything, and the CEO decides they need to “do AI.” So they hire a team of expensive data scientists, give them a blank check, and tell them to go find some magic in the data. A year later, all they have to show for it is a massive bill and a few science projects that have no connection to the actual business.
Feeling lost in the AI hype? You’re not alone. Here’s a practical, no-BS guide for non-technical leaders to successfully navigate an AI transformation.
Stop Chasing Ghosts and Start Solving Problems
The biggest mistake I see is leaders starting with the technology. They get obsessed with a specific tool or technique they read about and then try to force-fit it into their business. That’s completely backward.
Your job as a leader isn’t to understand the intricacies of a neural network. Your job is to understand your business and your customers. Where are the biggest points of friction? What are the most expensive, time-consuming processes? Where are you losing customers? Start there.
At RemoteTeam, we didn’t start by saying, “Let’s use AI.” We started by noticing that our onboarding process for new hires was a mess. It was manual, slow, and prone to errors. We were growing fast, and it was a huge bottleneck. The problem was clear: how can we make onboarding faster and more efficient?
Only then did we start exploring solutions. And it turned out that a simple rules-based automation system, with a little bit of natural language processing on top, could solve 80% of the problem. It wasn’t the sexiest AI on the planet, but it worked. It saved us hundreds of hours and made a real difference to the business. That’s the mindset you need.
My 5-Step Guide for Leading an AI Transformation
So, how do you do it? Here’s my playbook, forged from my own successes and failures.
1. Be a Detective, Not a Fortune Teller
Forget trying to predict the future of AI. Instead, put on your detective hat and go investigate your own company. Your goal is to find the most promising clues—the business problems that are practically begging for an AI solution.
- Follow the money: Where are your biggest costs? Is it customer support? Manual data entry? Inventory management? These are all prime candidates for AI.
- Listen to the complaints: What are your customers and employees complaining about? What’s making their lives harder? I once had a sales team that was spending half their day manually logging calls. That’s a massive, expensive problem just waiting to be solved.
- Look for the data trails: Do you have a lot of data that’s just sitting there, collecting dust? Customer feedback, support tickets, website analytics—it’s all potential fuel for an AI engine.
2. Assemble Your A-Team (Hint: It’s Not Who You Think)
You don’t need a team of 10 PhDs from Google. In fact, that can be a recipe for disaster. You need a small, scrappy team of problem-solvers.
Here’s who I look for:
- The Business Expert: Someone who knows your business inside and out. They understand the customers, the processes, and the politics. This person’s job is to keep the team focused on solving real problems, not just playing with cool tech.
- The Data Geek: You do need someone who can wrangle data. But they don’t need to be a world-renowned expert. I’d rather have someone who’s curious, resourceful, and knows how to get their hands dirty with SQL and Python than a theorist who only wants to work with perfectly clean data.
- The Engineer: Someone who can actually build things. A good software engineer who’s not afraid to learn new things is worth their weight in gold. They’ll be the one to turn the data geek’s models into a real product.
3. Your First Mission: The Pilot Project
Don’t try to boil the ocean. Your first AI project should be a small, well-defined pilot. The goal here is not to change the world, but to get a quick win, learn a ton, and build momentum.
At MovieLaLa, we wanted to improve our movie recommendations. Instead of trying to build a brand new recommendation engine from scratch, we started with a tiny pilot. We took a small segment of our users and just tried to predict whether they would like a single movie. It was a simple, focused problem. We built a basic model, and it worked. It wasn’t perfect, but it was better than what we had. That gave us the confidence—and the data—to go bigger.
Your pilot project should have a clear, measurable goal. Something like “reduce customer support response times by 10%” or “increase the click-through rate on our product recommendations by 5%.”
4. Measure Twice, Cut Once
How do you know if your AI project is actually working? You need to measure it. And I’m not talking about fuzzy metrics like “user engagement.” I’m talking about cold, hard numbers that connect to the bottom line.
Before you even start your pilot, you need to define your success metrics. What will success look like? How will you measure it? And how will you know if the AI is actually responsible for the change?
This is where having a business expert on your team is so important. They’ll help you focus on the metrics that actually matter to the business, not just the ones that make the data scientists feel good.
5. Scale It Up (and Don’t Be Afraid to Screw Up)
Once you’ve got a successful pilot under your belt, it’s time to scale up. This is where things can get tricky. What worked for a small pilot might not work for the entire company. You’ll need to think about things like infrastructure, data pipelines, and how to integrate the AI into your existing workflows.
But the most important thing is to not be afraid to fail. You will have projects that don’t work out. Models that don’t perform as well as you hoped. That’s okay. The key is to learn from your failures, kill the projects that aren’t working, and double down on the ones that are.
The Human in the Machine
It’s easy to get caught up in the technology of AI. But at the end of the day, AI is just a tool. It’s a powerful tool, for sure, but it’s a tool nonetheless. And like any tool, it’s only as good as the person wielding it.
Your job as a leader is to create an environment where people can use that tool to do their best work. That means giving them the freedom to experiment, the resources they need to succeed, and the psychological safety to fail. It means focusing on the problems, not the solutions. And it means remembering that the goal is not to “do AI,” but to build a better business.
So, stop reading the hype and start talking to your customers. The future of your company depends on it.
Frequently Asked Questions
What tools do I need to get started?
Start with the basics. You don't need expensive software or fancy tools. A spreadsheet, a note-taking app, and direct access to your customers will get you further than any enterprise platform. Add tools only when you hit a specific bottleneck.
How do I measure success with this approach?
Pick one or two metrics that directly tie to your goal and track them weekly. Vanity metrics like page views or follower counts rarely matter. Focus on metrics that reflect real engagement or revenue impact.
Do I need technical skills to lead an ai transformation (even if you don't understand the tech)?
Not necessarily. While technical understanding helps, the most important skills are clear thinking and the ability to break problems into smaller pieces. Many successful founders I've invested in started with zero technical background and either learned enough to be dangerous or found the right technical partner.