Designing a lightweight product delivery process

Orbit Partners, a boutique expert network serving private equity, came to me with a familiar problem. Like many lean teams, they had more requests for technical change than they could action, a single developer to do them, and no shared way to decide what was actually worth building.

I designed a delivery process for those constraints, adapted from the fixed-time, variable-scope approach I had built at AlphaSights (itself loosely based on Basecamp's Shape Up). The core idea is simple: work in fixed eight-week cycles, two weeks of reflection and planning and six weeks of execution, and treat time as fixed while letting scope flex, so something valuable always ships at the end of a cycle.

To decide what to work on, I added a lightweight request-and-prioritisation flow. Anyone can raise a request, framed as a problem to solve rather than a ready-made solution, with rough estimates of reach, impact and confidence. A simple RICE score (reach × impact × confidence ÷ effort) gives a first-pass ordering, as guidance and not gospel, so the team points its limited research time at the requests most likely to pay off. Chosen requests are then worked up into concrete, time-budgeted proposals, each with clear in-scope and out-of-scope boundaries and a plan for how scope gets trimmed if the deadline comes under pressure.

The process keeps stakeholders involved from the first conversation through to testing, so the people who raised a problem help shape and check the solution. Every cycle ends with a retrospective and a follow-up date, so the team actually measures whether a change delivered the value it promised.

The result is a repeatable, low-overhead way for a small team to aim scarce development time at the highest-value work, and to show the rest of the company visible, predictable progress.