Throughput
Throughput is the count of work items completed per unit of time (typically per week or per sprint). Unlike velocity (which is points-based and team-specific), throughput uses raw story count and is comparable across teams of similar story-sizing discipline.
Throughput is a sibling metric to lead time — both come from Little's Law: throughput × lead time = work-in-progress. Improving throughput without reducing scope means more parallel work, which often increases lead time. The healthier move is splitting stories smaller (each story → more throughput) and reducing context switches (lower WIP per engineer). Most teams chronically over-WIP by 2-3x what's healthy.
Long-form posts that explore throughput in depth — when to use it, common failure modes, how AI helps.
- BPMN process mining without Celonis moneyCelonis charges $100K-$1M+ for process mining. It's genuinely good. It's also wildly overpriced for 95% of teams. This is the lighter-weight playbook that actually works.9 min read
- What's the actual ROI of AI in software delivery?$4-$8 back for every dollar spent within 6 months, for most teams. The honest math from real data, not the deck.7 min read
Related terms
- Lead time
Lead time is the elapsed time from when work is requested (story created, ticket filed) to when it's delivered (deployed to production).
- Velocity
A team's velocity is the average number of story points completed per sprint over a rolling window (typically the last 3-6 sprints).
- Capacity planning
Capacity planning is the practice of estimating how much work a team can realistically take on in a sprint, accounting for PTO, meetings, on-call duty, and other non-coding time.