About this site
Better questions, practical changes and an honest assessment of the result. This site helps people investigate why improvement has stalled, choose a workable response and check what happens next.
A practical resource to help people plot data, recognise possible changes, ask better questions and learn from improvement work. It is not a substitute for professional judgement or specialist analysis when important decisions depend on the result.
Start with three questions
- Right problem? What is happening, and what conditions help explain it?
- Right intervention? What could we change—and what else is needed to make it work?
- Right measures? What evidence would help us judge the result and possible harm?
Start wherever you are most uncertain. The supporting guides introduce systems thinking, investigation and practical tools when useful—not as a compulsory sequence.
Explore the three questions →Who built it
Syd Stewart is a Chartered Chemical Engineer and founder of 4S Information Systems / DAWN Clinical Software. His background spans chemical manufacturing, plant management, information systems and more than forty years in clinical software supporting the care of people taking high-risk medications.
The site brings together that practical experience with a continuing question: how do we distinguish a useful improvement from ordinary variation—and understand what to do when the result is disappointing?
A little more about the background
Syd’s earlier work included research and development, production and plant management at ICI; business applications and data planning; and tutoring the Open University’s systems performance course. He founded the clinical software business in 1984, working with healthcare organisations in the UK and internationally.
The approach draws on Deming, Joiner, Goldratt, Senge, Meadows and the Model for Improvement, among others. These are sources of useful questions and methods—not guarantees that a proposed explanation is correct.
A free tool—with clear limits
The StepChange Analyzer helps explore time-series data and possible performance changes. It is free, requires no login and runs the analysis in your browser; the data you analyse is not uploaded. The website also uses page analytics.
A detected change is not proof that an intervention caused it. No detected change is not automatically proof of no benefit. Data quality, timing, variation, implementation and other explanations matter. Use the tool alongside knowledge of the work and appropriate professional and safety processes.
The examples cover healthcare, industry, public services and policy. They illustrate questions to investigate—not a promise that one statistical method suits every dataset.
Open the StepChange Analyzer →Understand the results and their limits →
Feedback and corrections
Questions, corrections and alternative explanations are welcome. If you spot a problem, identify the page, the claim and any relevant evidence. Suggestions from people applying these ideas in their own work are particularly useful.
Share feedback or a question →