The Roulette Principle: Why Cross-Industry Partners Win

TL;DR for innovation departments: Partnerships that reshape categories almost never come from within the same industry. Cross-industry partners combine non-overlapping assets, which is what makes the combination categorically new rather than incrementally better. This post sets out the Roulette Principle and why it should reframe how corporate innovation teams source and evaluate partnership candidates.

There is a peculiar phenomenon in the world of business partnerships. The ones that generate headlines, create viral moments, and unlock genuinely new markets almost never come from within the same industry.

They come from somewhere nobody expected.

Call it the Roulette Principle: the further apart two companies are on the conventional industry map, the higher the potential energy of their collaboration. Proximity breeds incremental improvement. Distance breeds revolution.

The Evidence Is Embarrassingly Clear

Consider Stanley and LoveShackFancy. Stanley makes insulated tumblers; LoveShackFancy makes romantic, floral-print clothing for women who shop at Bergdorf Goodman. Nobody at either company's board meeting would have called this an obvious partnership.

The limited-edition tumblers sold out within hours. Both brands' social media engagement spiked. The collaboration turned a functional product into a collectible moment, and a fashion label into a lifestyle ecosystem player.

Then there is Nike and SKIMS. Instead of a typical celebrity collaboration, they built a joint label: NikeSKIMS. Nike brought performance technology developed over 50 years. SKIMS brought design sensibility and a loyal female audience that Nike had consistently failed to capture on its own. The result was something neither company could have created alone.

Or look at what is happening in manufacturing right now. NVIDIA, a company that makes graphics chips, is partnering with ABB Robotics and Universal Robots to define "the factories of the future." NVIDIA's founders did not start a robotics company. They started a gaming company. Thirty years later, their chips are running the industrial automation revolution because nobody told them to stay in their lane.

Why Obvious Partnerships Disappoint

The logic of same-industry partnerships seems sound on paper. Shared infrastructure. Shared language. Shared customer base. Natural distribution synergies. Tick all the boxes.

In practice, same-industry partnerships tend to produce one of two outcomes: a collaboration that neither party is particularly committed to (because it mostly serves defensive purposes), or a drawn-out negotiation that collapses under the weight of competitive concerns, overlapping sales teams, and lawyers arguing about territory for six months.

The problem is that when two companies from the same sector work together, they are optimising for incrementalism. They know each other's constraints too well. They are solving known problems with known tools.

Cross-industry partnerships have the opposite dynamic. Both parties bring something the other genuinely does not have. There is no competitive tension, because the companies are not in the same game. And the combination of genuinely different capabilities tends to produce genuinely different outcomes.

The Discovery Problem

Here is where most businesses get stuck: even if you accept the logic of cross-industry collaboration, how do you actually find the right partner outside your sector?

Your professional network is biased towards people like you. Your industry conferences are full of your competitors. Your team's expertise is calibrated to your vertical. The further you try to look from your own industry, the harder it is to know where to look, what to look for, or whether what you find is actually useful.

This is not a hypothetical problem. YC-backed ArzuleAI was built specifically to solve it for B2B SaaS companies: using AI to identify which partnerships will actually drive revenue, rather than which ones feel comfortable. The fact that a company raised funding to build this in 2026 tells you that the problem is real and that existing tools are not solving it.

The Asset You Are Sitting On

Every company has an asset that looks ordinary from the inside and extraordinary from the outside. Stanley had a cult product that was functionally irrelevant to fashion. NVIDIA had chip architecture that gaming companies needed. Your company has something like this too.

The question is not whether you have a cross-industry partnership opportunity. The question is whether you are looking in the right direction to find it.

The Roulette Principle suggests the answer is: look somewhere you would not normally look. Not at your obvious adjacencies. Not at the companies you already know. Further out, where the combination of your capability and someone else's completely different capability could create something that did not exist before.

The algorithm, as it turns out, is remarkably good at pointing you there.

---

Bart Collet is the founder of Hyperadvancer, an innovation consultancy that turns unexpected partnerships into concrete business strategies. He is also the creator of the Ecosystem Innovation Roulette, an AI-powered tool that matches your company with an unexpected cross-industry partner and generates a concrete collaboration proposal. Spin it once. The algorithm has already decided.