Thinking on sprint planning, delivery forecasting, and the information problems that cost engineering teams the most.
We analysed delivery patterns across 12 simulated team archetypes. The failure mode is almost always the same: scope commitment without a simulation layer.
A 68% completion probability is not a prediction — it's a risk tolerance question. Here's how to use it in a planning conversation without losing the room.
Most sprint updates are written for the people who already know what happened. Here's how to write one that actually changes a decision.
Tech debt severity, ticket size, and team seniority are all present in your backlog. Most teams never surface them as a delivery risk until it's too late.
Adding a high-priority ticket mid-sprint is a negotiation, not a decision. Here's how to model the trade-off in three minutes so you go in with numbers.
Red, yellow, green — three states that encode everything the simulation found. How we designed a signal that collapses a 1,000-run simulation into one decision.
We publish roughly twice a month. No newsletters about newsletters.