How Zenlytic Avoided Being Just Another “AI Wrapper”: Lessons in Building Sustainable AI Products
The graveyard of AI startups is littered with companies that simply wrapped existing AI APIs in a new interface. In a recent episode of Category Visionaries, Zenlytic founder Ryan Janssen revealed how his company took a fundamentally different approach to building AI products.
The Danger of Being a “Thin Wrapper”
When ChatGPT launched in December 2022, countless startups rushed to build products around OpenAI’s API. But Ryan had already witnessed the dangers of this approach: “There are a bunch of tools that were doing AI web browsing… and then three days later, which is like 18 months in AI dog years, OpenAI launches the web browser capability and it just makes the startups all completely obsolete.”
Starting with the Problem, Not the Technology
Instead of starting with AI capabilities and looking for applications, Zenlytic began with a deep understanding of business challenges. Through their data science consultancy, they observed a fundamental problem: “Every company has more data than ever before, but nobody’s really using it to its full capacity,” Ryan explains.
This insight led them to focus on making data analysis more accessible. The goal wasn’t to build an AI company – it was to solve a persistent business problem. As Ryan puts it, “The real way to differentiate is not to be an AI business at all. The real way to differentiate is to solve a problem that happens to be using AI.”
Building Deep Technical Integration
Zenlytic’s approach to AI integration goes beyond surface-level applications. Rather than simply converting natural language to SQL queries – a common approach that Ryan notes is “probably accurate, like about 90% of the time” – they built a more robust solution.
“When you’re doing data analytics and you’re building board reporting off of this, 90% of the time is absolutely terrible,” Ryan emphasizes. Their solution involves integrating AI capabilities with a semantic layer that ensures consistency and reliability.
Demonstrating Reliability Through Action
In an industry filled with bold claims about AI capabilities, Zenlytic took a different approach to building trust. “Our objective is really to get to a demo in pretty much sort of every sales call,” Ryan shares. “We want to make it as easy as possible to show you that this works.”
This focus on demonstration rather than declaration helps overcome the skepticism many potential customers have about AI-powered tools. By showing rather than telling, they’ve been able to build trust in their technology’s reliability.
Staying Ahead of Rapid Change
The pace of AI development presents unique challenges for startups. “The pace of this change is like nothing I’ve ever seen,” Ryan notes. “We’re seeing stuff happen day by day with AI.” This rapid evolution requires companies to maintain a delicate balance between staying current with technological developments and building sustainable value.
Lessons for AI Founders
Zenlytic’s approach offers valuable lessons for founders building AI-powered products:
- Start with real business problems rather than AI capabilities
- Build deep technical integrations that can’t be easily replicated
- Focus on reliability and consistency over flashy features
- Demonstrate value through hands-on experience
- Maintain flexibility while building sustainable advantages
The key is remembering that while AI capabilities may evolve rapidly, the fundamental principles of building valuable products remain constant. Success comes from understanding your market, solving real problems, and building sustainable competitive advantages that go beyond simply wrapping the latest AI capabilities in a new interface.
As the AI industry continues to evolve, the companies that survive and thrive will be those that create genuine value through deep technical integration and focused problem-solving, not those that simply ride the wave of AI hype.
Twitter
facebook
linkedin