6 Key Lessons From Building a Global Fraud Platform From South Africa

The first customer, the raise, the investors, the team and the expansion: what one SA founder did differently at each stage.

Most of it runs against the usual startup advice, which is exactly why it’s worth reading.

We asked Orca Fraud co-founder and CEO Thalia Pillay, who has raised two oversubscribed rounds, what she learned getting here.

Lessons from building a global fraud platform from South Africa

1. Don’t chase your first customer, choose a design partner

When a LinkedIn post blew up and the inbound arrived, Thalia turned people away because she was looking for the right design partner: an early customer who helps you build the product before it exists. Being selective was a signal in itself, and the people she turned away became clients two years later.

Here’s the design partner approach.

2. Pick that partner on three questions

Her filter was simple. Is the client representative of their industry? Do they have real time to give feedback? Have they seen the world’s best tools? That last one is why her first design partner was the ex-Monzo fraud team, whose CPO later became an angel investor.

Here’s how to choose the right design partner.

3. Run the raise as a process, not a series of asks

Both rounds were oversubscribed, and she was blunt that it wasn’t luck. You drive hype, create FOMO, and ask people for advice instead of money. In practice that means batching first meetings into a few weeks and setting a real deadline, so investors feel each other in the room.

Here’s how to run a fundraising process that creates FOMO.

4. Give every investor a job

She built the cap table on purpose: a growth-stage fund as lead to learn from, a community of local angels for fast follow-on capital, and founders of decacorns in the fraud space to de-risk the company. Each seat was there for a reason beyond the money.

Here’s how to choose investors who bring more than money.

5. Stay lean, and raise less than you’re offered

Orca’s pre-seed of around $550k kept a two-person team shipping code and selling. The team is still about ten people, each covering roughly seven markets, made possible by AI. They could have raised two to three times more for the seed and chose not to, to keep their valuation low and avoid being diluted.

Here’s how to stay capital-efficient with a small, AI-leveraged team.

6. Reuse what works in one market to crack the next

After interviewing about 150 fraud analysts across Africa, they built models they could carry from market to market: mobile money in Kenya, then Pix in Brazil almost overnight, then PayShap in South Africa. Every market adds data nobody else has, and that data became the moat.

Here’s how to expand into new markets without starting from scratch.

The big payoff

Put together, it’s a company built by being fussy: about the first customer, the investors, the hires and the money. That discipline is how two founders with a hypothesis became a platform screening around $5bn a month in four years.