How to stand out when building an AI-powered SaaS with Trygve Karper @Databutton

saas.unbound is a podcast for and about founders who are working on scaling inspiring products that people love, brought to you by https://saas.group/, a serial acquirer of B2B SaaS companies.

In episode #24 of season 5 of the saas.unbound podcast, Anna Nadeina sits down with Trygve Karper, co-founder of Databutton. Databutton is an AI-powered no-code platform that enables users to build data-driven and AI-driven web applications simply through conversation. This insightful episode dives deep into Trygve’s journey, the vision behind Databutton, the challenges and opportunities in the AI SaaS space, and practical advice for founders looking to leverage AI in their products.

From Mathematics to AI-Powered No-Code Solutions

Trygve’s path to founding Databutton is as unique as it is inspiring. With a PhD in mathematics and years of research experience, he transitioned into AI, data science, and machine learning, joining the Norwegian scale-up Cognite as an early employee. There, he helped build AI and data science capabilities that supported digital transformation for large enterprises like BP and Exxon.

However, Trygve recognized a critical gap: many large-scale digital initiatives failed to empower employees to solve their own day-to-day problems. This insight became the foundation of Databutton’s mission—to enable individuals, especially non-technical users, to build custom tools and automate workflows themselves, using AI as an enabler.

What Is Databutton? Explained Simply

At its core, Databutton is an online AI agent that lets users describe the problems they want to solve and how they want to work differently. The AI then generates the full application stack—from UI to backend—allowing users to test, iterate, and deploy their apps seamlessly. This conversational approach removes much of the traditional complexity involved in building software.

Real-World Use Cases

• Mortgage Business Automation: A small Texas-based mortgage broker uses Databutton to automate the entire process from lead generation to deal closure, including payment history analysis, property appraisal, loan valuation, and marketing to investors.
• Sales Team Workflow Automation: A sales leader created an app to automatically upload PDFs into Salesforce, replacing tedious manual data entry.

These examples highlight Databutton’s flexibility—not just for SaaS product development but for automating a wide range of business workflows.

Addressing the Learning Curve and Technical Aspects

While Databutton is marketed as a no-code platform, it is built on a Python-centric architecture that generates React frontends and Python FastAPI backends. However, users do not need to write code themselves. Trygve acknowledges there is a learning curve, especially for first-time users, but emphasizes that no coding knowledge is required to create functional, production-ready applications.

With dedicated in-product support and a chat feature, Databutton helps users overcome hurdles quickly. Typically, building a fully integrated SaaS tool takes around three weeks for a non-technical user, which is comparable to other no-code tools but with the added benefit of full backend capabilities.

Pricing and Support Model

Databutton operates on a monthly subscription model, providing a one-stop solution with no hidden costs. Pricing tiers start at $200 per month for lighter use, with a $700 plan offering extensive support, including direct assistance in fixing issues and accelerating progress. This model ensures users get value and help throughout their development journey.

Overcoming the Blank Canvas Problem

One of the key challenges Trygve highlights is the “blank canvas” problem—users often struggle to envision the full scope of an app they want to build. To tackle this, Databutton encourages starting with a clear plan and breaking development into manageable tasks, iterating step-by-step. They also provide prompt guides and templates to help users communicate effectively with the AI agent.

The Reality of AI Adoption and Its Impact on Work

Trygve offers a balanced perspective on AI’s role in the workforce. AI excels at producing “good enough” results quickly—whether in writing, coding, or creative tasks—but it is not perfect. He points out that companies like Klarna faced challenges when replacing human roles with AI because AI struggles with nuanced human relations.

At the enterprise level, many organizations have yet to fully leverage AI productivity gains, sometimes resorting to layoffs instead of growth. In contrast, smaller companies and startups can use AI to compete more effectively, leveling the playing field by gaining access to capabilities that previously required large teams or budgets.

AI Tools in Databutton’s Development

The Databutton team uses a variety of AI tools such as OpenAI’s models, Anopic, Gemini, and chat-based interfaces to build and improve their product. Trygve notes that while some developers may use code generation tools like Cursor, the company’s main focus is on integrating AI to enhance user productivity and streamline internal operations.

Competition and Differentiation in the AI SaaS Landscape

Trygve positions Databutton as distinct from other no-code and AI app builders like Lobe, Bolt, and Devin. While many competitors cater to technical users focused on prototyping and MVPs, Databutton targets small business owners and non-technical users who want to create full production applications with real backend integration.

Key differentiators include:

• Full-stack app generation with Python backend
• Comprehensive user support to ensure success
• Focus on solving real workplace problems, not just quick prototypes

Growth Strategy and Customer Acquisition

The company’s main growth channels are YouTube and word of mouth. YouTube serves as a news source and educational platform for potential users, while referrals help build a loyal customer base. Trygve admits that while word of mouth is difficult to start, being part of the broader AI movement and having a differentiated product has helped them gain traction.

Validating Product-Market Fit Through Early Sales

Before fully launching, the Databutton team reached out proactively to potential users on LinkedIn, demoed the product, and secured paid subscriptions on the spot. While the conversion rate from initial outreach was low, those who engaged converted at around 40%, providing strong validation that paying customers found real value in the product.

Biggest Wins and Lessons Learned

Trygve’s biggest win was the board’s eventual belief in AI-generated full-stack app development. Initially met with skepticism, the success of competitors like Bolt and Lobe helped validate Databutton’s vision. On the flip side, the team acknowledges that marketing and sales are their weaker areas, coming from a strong product development background but still learning how to scale growth effectively.

A Hack for Balancing Work and Life as a Founder

To maintain balance, Trygve sets aside dedicated time daily with his children, treating it as a sacred commitment before diving into work. He also avoids micromanagement, empowering his team to own their results, which fosters excitement and ownership across the company.