Skip to content

FLOW LAB · FREE INTERACTIVE TOOL

Make space. See work flow.

Try a smaller work limit, a faster review, or fewer blockers. Watch the same jobs move through two pipelines and see what changes.

A toy model.
A useful conversation.
Day 0 / 60Paused
Start with a question
Shared stream: 2 jobs / day · 60 days · seed 711

One seed generates the same job sizes and blocker draws in both pipelines. Changing a setting resets the view to day 0.

Paused at day 0. Play or step through the same jobs in both pipelines.

Baseline

The way it works today.

0completed
WIP 16 · capacity 2 / 2 / 2 · blockers 20%
Baseline settings
NOT STARTED0 in backlog
No queue yet
01

Discovery

0

0 / 2 worked today

    Understand the work

    02

    Build

    0

    0 / 2 worked today

      Make the change

      03

      Review

      0

      0 / 2 worked today

        Check and finish

        DONE · 0

        Finished work will appear here.

        Throughput
        0.0jobs / day
        Cycle time
        —days
        Lead time
        —days
        Current WIP
        0of 16
        Blocked
        0jobs

        0 admitted jobs waiting · 0 arrived · blocked work holds a WIP slot, but does not use a worker.

        Experiment

        What happens if…

        0completed
        WIP 6 · capacity 2 / 2 / 2 · blockers 20%
        Experiment settings
        NOT STARTED0 in backlog
        No queue yet
        01

        Discovery

        0

        0 / 2 worked today

          Understand the work

          02

          Build

          0

          0 / 2 worked today

            Make the change

            03

            Review

            0

            0 / 2 worked today

              Check and finish

              DONE · 0

              Finished work will appear here.

              Throughput
              0.0jobs / day
              Cycle time
              —days
              Lead time
              —days
              Current WIP
              0of 6
              Blocked
              0jobs

              0 admitted jobs waiting · 0 arrived · blocked work holds a WIP slot, but does not use a worker.

              Finished work, over time.

              The same incoming jobs. Two different ways of making space.

              BaselineExperiment
              Cumulative completed jobs through day 0Baseline: 0 completed jobs. Experiment: 0 completed jobs. Exact values are also in the comparison table below.0510Day 0Day 60
              Comparison at day 0
              Flow measureBaselineExperiment
              Completed jobs00
              Throughput (jobs / day)0.00.0
              Average cycle time (days)——
              Average lead time (days)——
              Current work in progress00
              Average work in progress0.00.0
              Jobs in queues00
              Blocked jobs00

              Cycle and lead time use completed jobs only. A dash means nothing has finished yet. Queue totals include backlog and admitted waiting jobs.

              Keep the experiment.

              Save the numbers, settings and seed so you can explain the comparison.

              Settings stay in this browser after your first edit.

              Exports cover day 0 through day 0. Select Complete run to export all simulated days.

              READ THE SYSTEM, NOT JUST THE SCORE

              More started
              isn’t always more finished.

              Work limits make queues visible. Extra build capacity can simply move a queue to review. A blocker can hold space even when workers are free. Change one setting, replay the same seed, and ask where the work waits.

              Read The Kanban Guide

              What this model assumes

              • New jobs arrive at the start of each day. Seeded work takes 1–3 days in Discovery, 2–5 in Build and 1–3 in Review.
              • A global WIP limit controls admission. Once admitted, a job keeps its slot through stage queues and blockers until it finishes.
              • Each available worker processes one unit of work per day. Jobs are handled in stage-entry order, then job number. Blocked jobs use no worker capacity. There is no multitasking or context-switching penalty.
              • A job has one seeded interruption check before it first receives work in Build. The blocker duration includes the day it begins.
              • Discovery and Build hand off on the following day. Review finishes at the end of its processing day.
              • Cycle time runs from admission to completion; lead time starts at arrival. Both include the start and end day. Throughput is completed jobs divided by elapsed days. Average WIP is the mean of end-of-day counts.

              This is a simplified simulation, not a forecast for your team. It leaves out dependencies, rework, priorities and human context. Use it to form a question, then check that question against real work.

              Your experiment stays with you.

              No account, real project data or uploads. Settings are stored only in this browser after an edit. The jobs and results are generated here from the seed. Exports are yours to keep and share.