What is Throughput and Why Should You Care?

In the world of flow metrics and value delivery, throughput is a simple but powerful concept. It can be a clear, shared metric for teams — and, more importantly, a way to spark meaningful conversations about how to improve.

In this post, we’ll break down what throughput is, why it matters, how to measure it, and some practical ways that teams can use it to work better together.

What is Throughput?

The Kanban Guide defines throughput as "the number of work items finished per unit of time."
‍
It is an exact count — not an average, not a range. 

Throughput answers:  How many <work items> got done in a given period of time?

It’s worth noting for throughput to be accurate and useful, teams must have a shared understanding of:

• What is being measured (stories, features, epics, widgets?)
  
  
• When something is considered finished (the "finish line" of your workflow, or a clear Definition of Done).
  

Without alignment on these two points, throughput measurements risk becoming misaligned or misleading.

Pro tip: The Kanban Guide’s section on Definition of Workflow offers great advice on defining your work items and finish points — both crucial precursors to reliable throughput measurement.

Why Measure Throughput?

Throughput is one of the four flow metrics (alongside Work-in-Progress <WIP>, cycle time, and work item age). Together, these metrics paint a fuller picture of how value flows through your system — and where bottlenecks, risks, or improvement opportunities might lie.

Some specific reasons to pay attention to throughput:

1. Keep a Focus on Delivery

It’s easy to get caught up in estimating, planning, and prioritizing — and forget about actually finishing work.
‍
Throughput reminds us to stop starting work and start finishing it — focusing on delivering small batches of value, collecting feedback, and adapting.  Throughput can help us keep our ‘eyes on the prize’ - with the “prize” being the feedback loop!

2. Understand Team and System Dynamics

Throughput is one indicator that might help us understand churn, disruption, gaps, impediments, and the like. It can be very revealing to look at a chart of throughput data over time, and layer in qualitative context such as organizational changes, onboarding, technology challenges, changes in priorities, etc. What patterns can you see? What hypotheses for improvement can you form?

3. Assess Workflow Balance

By comparing throughput to your workflow’s arrival rate (or backlog depth), you can assess the relative balance of your workload.
If incoming work consistently outpaces throughput, it’s a recipe for overload — and an invitation to find ways to improve your system’s stability.

4. Enable Probabilistic Forecasting

Throughput fuels probabilistic forecasting methods like Monte Carlo simulations — which can revolutionize how you answer the age-old question, "When will it be done?"
(Check out Drunk Agile Episode 3: Probabilistic Forecasting using Monte Carlo Simulation if you want more info on this topic!)

But Wait… What about ‘Outcomes > Output?’

Yep. 100%. Throughput is about output — and ultimately, we care about outcomes.
‍
But here’s the thing: We can’t actually assess outcomes until after you deliver something.

Throughput helps shorten the feedback loop by encouraging small, frequent deliveries. This enables faster validation of whether the thing you built is actually valuable — and provides more opportunities to adapt and course-correct.

We can’t measure outcomes without producing output. 

How to Measure Throughput

It’s refreshingly simple:

1. Define your finish line (what “done” means).
   
   
2. Track completed work items (using IDs or other identifiers) and their completion dates as they cross that finish line.
   
   
3. Select your time window.
   
   
4. Count how many items finished during that window.
   

That’s it. That’s your throughput.

Examples:

• Number of stories completed per sprint
  
  
• Number of bugs resolved per month
  
  
• Number of features delivered per quarter
  

Important: Throughput ≠ Velocity.

• Throughput counts finished work items — work items being the chunks of potential value moving through your workflow.
  
  
• Velocity counts story points  — story points being…. Let’s be honest, who actually knows??
  

Throughput as a Team Sport

Throughput is most valuable when it’s owned and discussed by the whole team.

Some ideas for making throughput a team sport:

• Collaborate actively: Ask "What can we do to complete this work item sooner?"
  
  
• Watch work item age: The older a work item gets, the less likely it is to complete within the desired timeframe.
  
  
• Review throughput together: Look for trends and discuss ways to stabilize or improve.
  
  
• Raise blockers early: Don’t wait to highlight risks to your flow of value!
  

Throughput isn’t about blaming or pressuring. It’s about building shared awareness and finding better ways to deliver value, together.

In Summary

Throughput measures how much work a team completes in a given time — but its real power lies in what it enables:

• Sharper focus on finishing
  
  
• Deeper insights into system health
  
  
• Stronger forecasting
  
  
• Faster feedback loops
  

When embraced thoughtfully, throughput becomes less about counting work and more about accelerating learning, improving flow, and delivering better outcomes!

‍