Every product on the Bison Hill site has been through the same process before it shipped. By 2025, that process has been refined through four years of development into something formal enough to be called a standard. Here is what it looks like.
The five stages
Stage one is the desk review. Before a single material is ordered or a prototype made, we ask whether the brief holds. Is the problem real? Is there a genuine gap that this product fills, or is it a solution looking for a problem? Most product ideas fail here, and that is the cheapest place for them to fail.
Stage two is the prototype test. Any prototype that survives the desk review goes to the fire. Minimum three hours of active cooking. Minimum three separate sessions. Not one long session — three distinct occasions, different cooks, different conditions. A single session can produce a false positive. Three sessions across different days produce honest data.
Stage three is the Grill test. The prototype goes to a real Bison Hill Grill event — a wedding, a festival, a private catering booking. It is used under pressure, with real guests, with no fallback option. If it fails here, it fails in front of people who expected it to work. That is an uncomfortable situation deliberately designed to be uncomfortable. Products that cannot handle real conditions should not ship to customers who will use them in real conditions.
Stage four is the stress test. What happens when it goes wrong? Drop test. Heat test. Force test. We are looking for failure modes, not confirmation that it works when handled carefully.
Stage five is team consensus. Everyone on the team who cooks signs off. Not a majority vote — unanimous sign-off. If one person who has used the product has a reservation, we resolve the reservation before production is authorised.
Why the filter exists
The minimum timeline from prototype to production sign-off is six weeks. We have stopped development on two products between 2024 and 2025 that failed stages three or four. Both were functional products. Neither was good enough.
The filter exists because shipping something that fails in the field is not a recoverable mistake for a small brand. A single product that disappoints a customer erodes the trust that every previous product built. The buy-once promise only means something if every product shipped is genuinely worth buying once.
