The criteria that separate a real design partner from a friendly waste of time, how many you need, and how to structure it.

Last time we explained what a design partner actually is, and why Thalia Pillay turned away paying customers to find the right one on her way to building a platform that now screens around $5bn a month. That was the what. This is the how.

Because the concept only works if you’re ruthless about who you let in. The wrong partner doesn’t just waste your time; they actively pull your product in a direction nobody else wants to buy.

Enthusiasm in a first meeting is cheap. What you’re looking for is someone whose problem is deep, urgent and visible enough that they’ll give up their own time to help you fix it. If the pain is shallow, you get politeness instead of progress, and polite feedback is worse than none because it feels like validation.

Thalia’s first question was whether a client is representative of their industry, and it’s the one that saves you from the most expensive mistake in this whole practice. Build around a partner whose needs are unusual and you end up with custom software that serves exactly one customer. Ask how closely their workflows and systems resemble the wider market you intend to sell to, and whether the person giving you feedback holds a role that exists at every other target company.

Her second question was whether they have the time and capacity to give real feedback, and this is where most partnerships quietly die. A useful test: if they won’t commit to a thirty-minute follow-up next week, the pain isn’t urgent enough to carry a six-month relationship. Capacity is political as well as practical, because a champion too junior to influence a buying decision can love your product without it ever converting into anything.

Her third question was whether they’ve seen the world’s best tools. Someone whose reference point is a spreadsheet will be delighted by something mediocre; someone who has used the sharpest software in their category knows exactly what good looks like and will hold you to it. That’s why Orca’s first design partner was the ex-Monzo fraud team, people who had worked inside one of the best fintechs in the world. Their former CPO is now an angel investor in the company.

One or two partners and you overfit to an edge case, mistaking one company’s quirks for market truth. More than ten and the feedback fragments into noise you can’t act on, with nobody getting enough of your attention to produce real signal. Three to six deeply engaged partners is the range that lets you spot patterns while staying hands-on, and it gives you a way to sanity-check requests: if one partner asks for something and the others don’t need it, that’s a custom build, not a core feature.

Agree up front what each side commits to: a regular feedback rhythm, a shared channel, and a written definition of what success looks like for them. Decide the money question early too. Free or heavily discounted during the build is normal, but define what triggers the shift to paid, whether that’s usage, time or proven value. When you hit it, invoice them. If they pay, you’ve proven the value is real; if they don’t, you’ve learned they were never the right partner. Both answers are worth having.

Choose well and your design partners do three jobs at once: they shape a product the wider market actually wants, they become your first paying customers, and their names become the proof that opens the next ten conversations. Choose badly and you’ve spent your scarcest runway building something for one company that was only ever being polite.

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.