Bets, Not Backlogs

Most software teams maintain a backlog — a massive, ever-growing list of features, bug fixes, and nice-to-haves they might get to someday. In practice, it functions as a graveyard. Most of the ideas in there are either out of date or weren't urgent enough to act on when they were added. The ones that were urgent got done — and the rest have been quietly losing relevance ever since. The backlog doesn't capture future value. It generates ambient guilt.

Shape Up, Basecamp's project management framework, offers a different model built on a simple premise: if an idea is truly important, it will keep coming back. You don't need to track it in a ticket.

Every six weeks, a small group decides what to do now. They look at the current state of the market, the team, and the product, and place a bet on a specific pitch. Ideas that don't get chosen don't carry over — they disappear. Every idea has to re-earn its place based on its current value, which keeps the team focused on what matters today rather than obligated to what seemed important last quarter.

The fixed cycle is where the model gets its teeth. A bet is six weeks, and when the time is up, the project ends. If the work isn't done, the team ships what they have or cuts scope to fit. No extensions. This does two things: it removes the sunk cost pressure that keeps dying projects alive, and it forces honest scoping decisions to happen at the beginning of a cycle rather than the end — which is where they belong. When teams know the deadline is real, they stop negotiating scope in the final week and start negotiating it on day one.

If an idea is worth pursuing after a cycle ends, it makes the case again. The discipline is in never letting how close a feature is to finished become a reason to continue.