Compliance is not a problem that can be solved with a scrappy prototype. That is the argument from Daniel Kjellén, CEO and co-founder of Freda, a Stockholm-based AI-native compliance company that emerged from stealth in September 2026. Kjellén previously co-founded the open banking platform Tink in 2012 and led it through its €1.8 billion acquisition by Visa in 2022. In a first-person account, he explains why the standard startup playbook of building a minimum viable product fails when the problem is compliance.

What You Need to Know

Compliance obligations are fragmented across legal, product, and operations teams, making piecemeal solutions ineffective. Traditional compliance startups target a single department with a clear budget, but this approach leaves other risk areas unaddressed. Kjellén argues that solving compliance requires a comprehensive platform from the start, not a narrow MVP. The rise of large language models has made building such broad systems feasible for small teams.

The Compliance Hydra

Kjellén describes compliance as a Hydra: cut off one head and two more grow back. Every company's obligations depend on countless factors spread across different parts of the business that rarely communicate. Legal tracks regulations, another department oversees implementation, product teams make feature decisions, and yet another team checks whether those features are compliant. Policies and standards belong to other departments entirely. A compliance startup typically builds for one silo, but that leaves the rest of the beast alive.

Why a Minimum Viable Product Fails

The lean startup methodology, popularized by Steve Blank and Eric Ries, encourages founders to build a minimal product to test assumptions and iterate quickly. For most software, this works. For compliance, however, the interconnected nature of obligations means that fixing one area does not reduce risk elsewhere. A product that addresses only one department's needs may satisfy a buyer but fails to solve the underlying problem. Kjellén contends that compliance requires a maximum viable product from day one, a full build that covers all relevant areas.

  • Legal: Tracks regulations and interprets their applicability.
  • Product: Makes feature decisions that may have compliance implications.
  • Operations: Implements and monitors compliance across the business.
  • Policies: Sets internal standards that must align with external rules.

The Second-Time Founder Advantage

Kjellén's experience with Tink gave him unique advantages when starting Freda. After selling Tink to Visa, he and co-founder Fredrik Hedberg had both capital and credibility. This allowed them to fund the full build themselves without investor pressure to shrink the vision. They could also attract talent and partners more easily. Stockholm's ecosystem, with alumni from Spotify and Klarna, has seen a wave of second-time founders tackling harder problems. Kjellén notes that experience and capital used to get you only so far, but language models have changed that. Freda's team of 22 ships at a pace that would have required hundreds of people a decade ago.

Why This Matters

The compliance technology market is growing as regulations multiply and businesses face increasing scrutiny. Startups that cling to the MVP approach may find themselves building point solutions that fail to address the root problem. Investors, meanwhile, are increasingly funding second-time founders who can deploy larger amounts of capital to build comprehensive platforms. This shift could lead to consolidation in the compliance sector, as broad platforms outcompete narrow tools. For founders, the lesson is clear: some problems demand total commitment from the start. For enterprises, the implication is that integrated compliance solutions may become the norm, reducing the need to stitch together multiple point products.

What You Need to Know

Compliance is not a feature that can be bolted on later. It is a core business function that touches every department. Startups that treat it as a narrow use case risk building products that fail to solve the real problem. The rise of AI and large language models has lowered the barrier to building comprehensive compliance platforms, making it possible for small teams to tackle what once required hundreds of employees.