The Product Oxymoron: Building for Chaos
Lessons from the front lines of Snapchat and Discord
Peter Sellis has spent a decade at the epicentre of consumer product mania. From the early days of Snapchat to the rapid scaling of Discord, he has seen how products actually take hold of the human psyche. His perspective is stripped of the usual corporate gloss. He does not talk about 'optimising user journeys' or 'leveraging synergies'. Instead, he talks about the messy, often violent reality of building things people actually use. The core problem in modern product management is that most people are too careful. They try to manage growth like a controlled experiment, forgetting that real growth is often a chaotic, unpredictable explosion that comes from the product itself, not from a marketing spreadsheet.
The Terrorist Organisation Model
One of Sellis's most striking ideas is his approach to team structure. He advocates for designing teams like terrorist organisations. This isn't about violence; it is about autonomy, speed, and a singular, uncompromising focus. In a standard corporate structure, teams are bogged down by consensus, layers of approval, and the need to please every stakeholder. A 'terrorist' team, by contrast, operates with a high degree of independence and a mission-driven intensity. They don't wait for permission to iterate. They strike at a problem, deploy a solution, and move on. This radical decentralisation is what allowed Snapchat to move at a speed that larger, more 'professional' competitors couldn't match.
Growth almost always comes from the core product, not the marketing machine.
The median product manager is often a failure because they focus on the wrong metrics. They obsess over incremental improvements to existing features rather than asking if the feature should exist at all. They become administrators of the status quo. Sellis argues that great product management requires a willingness to be wrong and a capacity to handle the friction that comes with true innovation. If you are only making things 1% better, you aren't building a product; you are managing a decline. The real work is finding the 'oxymorons'—the contradictory truths that drive a product forward, such as being both incredibly simple to use and incredibly powerful in function.
- Prioritise core product utility over marketing hype
- Build autonomous, mission-driven teams
- Avoid the trap of incrementalism
- Embrace the tension of product oxymorons
Ultimately, the difference between a product that fades and one that defines a generation is the courage to let the product lead. When Snapchat's ads business failed to reach its full potential, it wasn't because the sales team lacked skill; it was because the product's core identity was at odds with the traditional advertising model. To build something that lasts, you must protect the integrity of the user experience against the short-term demands of the revenue model. This is the hardest battle a product leader will ever fight.
Stop managing features and start building missions.