A little sustainability insight - Royd Brayshay - New Redo

A little sustainability insight - Royd Brayshay - New Redo

Royd Brayshay of New Redo discusses how knowledge of Little’s Law can improve your management decisions.

(This post is part of a collection from the Sustainable Digital Delivery event hosted by Conflux at the Leeds Digital Festival 2019)

Key Points:

  • We don’t use metrics enough to make management decisions

  • Little's Law: Cycle-time = WIP/ Throughput - comes from queuing theory

    • Limiting WIP positively improves cycle time, reducing cycle times (reduce blockages) reduces IP

    • Adding capacity (throughtput) with people or process improvement benefits cycle times

    • Assumptions: Average cycle time = Average WIP / Average Throughput

    • Note: given a stable system

      • consistent units

      • average time = average departure date

      • All work leaves the system done

      • average age of WIP is approximately constant

      • total amount of WIP is approximately constant

  • There is a mathematically provable relationship between the work coming in, and the work leaving.

Conflux acquires Skelton Thatcher Publications - new books for engineers

Conflux acquires Skelton Thatcher Publications - new books for engineers

Conflux has acquired Skelton Thatcher Publications, a pioneering publisher of books for engineers and practitioners in the software industry.

The Team Guide books from Skelton Thatcher Publications cover Software Operability, Metrics for Business Decisions, Software Testability, and Software Releasability, all areas that are often overlooked by organisations building software systems.

Don’t estimate: forecast! Predicting software delivery using metrics — Martin Aspeli

I was fortunate to hear a talk recently by Martin Aspeli, Head of Engineering at Deloitte Digital, explaining how and why to use metrics to predict software delivery rather than estimates. Here are my brief notes:

tl;dr: there is no need to spend time estimating — with discipline and data we can forecast!

  1. Predictability is more important than pure top speed for most organisations.
  2. Humans are notoriously bad at estimating. This is a facet of human nature — as a species we are bad at estimating. See Hofstadter’s Law.
  3. We tend to ignore Wait Time when estimating, yet Wait Time makes up the majority of Cycle Time in most organisations doing knowledge work like software development.
  4. With some discipline in how we use Jira, we can start to use metrics to help track how long things actually take — this increases predictability.
  5. We need to take care to ensure that the Arrival Rate of new work items does not exceed the Departure Rate. Treat our team and its context as a system to be cared for.
  6. Use a Cumulative Flow Diagram (CFD) to track the system health.
  7. Measure end-to-end Cycle Time of actual stories. No need to size the stories with Story Points or t-shirt sizes. The ensemble of stories provides confidence measures.
  8. As people are “loaded” beyond about 80% capacity, their capacity to respond drops dramatically. This is basic queueing theory.
  9. Forecast the delivery dates for batches of work using Monte Carlo simulations. This gives confidence measures (50%, 85%, 95%, etc.) allowing stakeholders to choose dates based on likelihood and risk.

Further reading