Why lean beats large, how AI changes the maths, and the analogy that explains where your money actually goes.

Capital efficiency isn’t a constraint you tolerate until you can afford to be wasteful. Done on purpose, it’s one of the most attractive things an investor sees.

We asked Orca Fraud co-founder and CEO Thalia Pillay, who runs roughly seven markets per person across a team of about ten, how she keeps it that lean…

Coming from payments, Thalia and her co-founder Karla knew margin is everything, so they built on strong unit economics from day one rather than treating profitability as a someday problem. Their instinct is conservative to the point that their own investors ask them to spend faster. The discipline isn’t stinginess; it’s that money raised is ownership sold, and most of it gets spent worse than you expect.

“It’s like a wedding. You buy a 200-rand cake, throw a wedding in front of it and it’s 2,000.”

Orca raised its pre-seed in March 2024, pre-idea and pre-revenue, off the back of AI. An initial sum of around $550k kept a two-person team shipping code and selling out of Thalia’s living room, and her verdict is unambiguous: without AI, without Claude, they wouldn’t have had those opportunities. The team is still only about ten people, mostly engineers, covering roughly seven markets each. AI is what makes that ratio possible, and it’s the difference between hiring fifty people to compete globally and doing it with a handful from South Africa.

The payments background shows up here. Rather than chasing growth and worrying about margin later, they engineered the business so the numbers worked from the start. It’s slower-feeling early on and it’s exactly what lets you stay small, because a business with healthy unit economics doesn’t need to raise and hire its way out of a hole it dug chasing scale.

They could have raised two to three times more for the seed and chose not to, on the principle of keeping your valuation as low as long as possible. The questions Thalia asks of any raise: does this valuation make sense, can I grow into it, and will it create pressure I don’t want. Early on they went out for R100k and an investor called it an incredible amount; she’s glad they didn’t take a bigger cheque, because they’d have been, in her words, hella diluted.

The wedding-cake problem is that expenses balloon the moment you stop watching them, and enterprise sales in particular is slower and more expensive than anyone expects, so keep a buffer for the rainy day that comes. Where the money should go is people: building and sustaining world-class talent is the best thing she invests in. The unglamorous rest matters too. Orca runs a US Delaware company alongside an SA one, and understanding the tax structure now takes up a real chunk of her time, precisely because getting it right early saves far more later.

This one cuts against what every founder is currently doing. Thalia deliberately avoids shouting that Orca uses AI, because her favourite tools are the ones that don’t: the same way you can use a crypto company’s product without needing to know there are stablecoins on the back end. AI is how you stay lean, not the pitch. Leading with “AI-powered” dates you and tells customers nothing about the outcome they actually care about.

Stay lean on purpose and two things follow. You keep more of your company because you raised less to get further, and you become exactly the kind of business investors chase, because capital efficiency reads as a team that will do a lot with whatever you give them.

Thalia’s full talk from Founder Collab Live is available to members inside the Founder Collab, where she goes further than we could cover here:

You’ll also get access to 40+ masterclasses from SA founders and operators on sales, fundraising, UX, paid media and more inside The Founder Collab.

Get more SA tech and business news, tips and business-building workflows and subscribe to The Open Letter.