Swarm pattern
Swarming is the practice of having multiple team members work on the same story until it's done, then move together to the next. Swarming maximises throughput (one story done in a day beats five stories partially done in a week) and reduces WIP.
Swarming sits between solo work and full mob programming. Two-to-three engineers on the same story, possibly with a PM available for AC clarifications, often delivers in a day what would take 3-5 days of parallel solo work. The cost: feels less efficient ('three engineers on one story?') even when throughput data says otherwise. Healthy adoption pattern: swarm on high-priority and high-uncertainty work; solo on routine work.
Related terms
- Pair programming
Pair programming has two engineers at one workstation, alternating between driver (typing) and navigator (reviewing, suggesting, thinking ahead).
- Mob programming
Mob programming is a practice where the entire team works on the same problem at the same time, on the same screen, with one person typing (the 'driver') and the rest navigating.
- Throughput
Throughput is the count of work items completed per unit of time (typically per week or per sprint).