Experimental Innovation Cycle

A tool to systematize the process of innovation through structured experimentation in business contexts.

Builds on the PDSA cycle and the Model for Improvement (Walter Shewhart, W. Edwards Deming, and Gerald Langley and colleagues, The Improvement Guide, 1996), the business-experiment discipline of Stefan Thomke (Experimentation Matters, 2003) and Thomas H. Davenport ("How to Design Smart Business Experiments," Harvard Business Review, 2009), and Eric Ries's build-measure-learn loop (The Lean Startup, 2011).

What it is for

The purpose of the Experimental Innovation Cycle thinking tool is to systematize the process of innovation through structured experimentation in business contexts. It enables businesses to approach problem-solving with a mindset that views failures and setbacks as valuable learning opportunities. Each step in the tool is designed to guide businesses through a cycle of envisioning innovation, formulating hypotheses, designing and executing experiments, and analyzing results. This cycle not only fosters resilience and adaptability but also encourages a culture of continuous improvement and strategic growth. By iteratively refining ideas and approaches based on real-world feedback and insights, businesses can develop innovative solutions and adapt more effectively to an ever-evolving market landscape. This tool is particularly useful for businesses seeking to foster a culture of innovation and for those looking to navigate complex challenges with a methodical, data-driven approach.

Why use it

In business problem-solving, viewing experimentation and failure as learning opportunities is vital. Each failure provides new insights, debunking assumptions and revealing unforeseen paths, much like a tech startup refining its software through testing. This approach fosters resilience and innovative thinking, transforming setbacks into strategic growth opportunities. Embracing this mindset allows businesses to adapt and thrive in an ever-changing landscape, making failures stepping stones to success.

How to use it

Define Innovation Goal. Determine the specific innovation you aim to explore in the experiment.
Set Learning Goals. Define what you want to learn or understand from conducting the experiment.
Formulate Hypothesis. Formulate your expected outcome or result of the experiment.
Plan Experiment. Outline the steps of the experiment, ensuring simplicity for effectiveness.
Conduct Test and Evaluation. Compare the actual outcomes of the test with your initial hypothesis.
Extract Insights. Highlight key learnings and insights gained from the experiment.
Determine Next Actions. Decide the subsequent action or experiment, making minimal changes for speed.

The seven steps are the short form; the worksheet below runs the same cycle in eleven stages, each with its key question and the outcomes it should deliver.

The eleven stages of an experiment

From the vision of the innovation to the feedback loop that improves the next experiment. Select a stage to see what it involves, the questions it answers, and the outcomes it can deliver.

Vision of Innovation

Key questions: What is our end goal with this innovation?
Description: Define the overarching innovation goal, including the end-state vision of what the experiment aims to achieve or explore.
Potential outcomes: New innovative products/services; breakthrough business models.

Learning Agenda

Key questions: What specific knowledge or skills are we aiming to acquire?
Description: Develop a learning plan detailing the key knowledge areas, skills, or insights the experiment seeks to address or develop.
Potential outcomes: Enhanced skills; deeper market understanding; strategic insights.

Hypothesis Formulation

Key questions: What are our expected outcomes based on our current understanding?
Description: Construct a clear hypothesis or set of predictions based on current knowledge and assumptions.
Potential outcomes: Validated/invalidated assumptions; new hypotheses for future experiments.

Experiment Design

Key questions: How will we conduct the experiment? What resources are needed?
Description: Craft a detailed plan for the experiment, including methodology, tools, and timeline. Incorporate feasibility considerations.
Potential outcomes: Successful execution of the experiment; identification of operational challenges.

Resource Allocation

Key questions: What resources can we allocate to this experiment?
Description: Determine the resources required for the experiment, including budget, personnel, technology, and other assets.
Potential outcomes: Optimal resource utilization; cost efficiency; minimized resource wastage.

Risk Assessment and Mitigation Strategy

Key questions: What are the potential risks and how can we mitigate them?
Description: Identify potential risks and devise strategies to mitigate them. This step is crucial for reducing potential setbacks.
Potential outcomes: Reduced impact of risks; successful navigation of challenges.

Implementation and Iteration

Key questions: How can we adapt our approach based on real-time data?
Description: Execute the experiment, ensuring flexibility for adjustments and iterations based on ongoing findings.
Potential outcomes: Iterative improvement; refined experimental process.

Data Collection and Analysis

Key questions: How does the data align with our hypothesis?
Description: Gather data systematically during the experiment. Analyze the results in the context of the hypothesis and learning agenda.
Potential outcomes: Data-driven decision-making; actionable insights.

Insight Synthesis

Key questions: What have we learned from this experiment?
Description: Extract key insights and learnings from the experiment. This involves interpreting results and integrating them into knowledge.
Potential outcomes: Strategic knowledge; basis for future decision-making.

Application and Integration

Key questions: How can we apply these insights in our business model?
Description: Apply the newfound insights to relevant business areas. Integrate them into practices, products, strategies, or operations.
Potential outcomes: Improved business processes; enhanced product/service offerings.

Feedback Loop and Continuous Improvement

Key questions: How can we optimize this process for future experiments?
Description: Establish a feedback mechanism to continually refine and improve the experimentation process based on outcomes and insights.
Potential outcomes: Continuous improvement; sustained innovation and growth.

Where it fits

The process recommends it at Designing and Testing the Solution, the first step of the Execute phase, where it is the one tool named.

Related tools

Within this series, the Blueprint Optimization Framework refines a single idea into a workable plan, where this cycle tests a prediction about it.

Design of Experiments (Fisher, Box, Hunter, and others)
PDCA Cycle (Shewhart and Deming)
Randomized Controlled Trial (Bacon, Mill, Fisher, and others)
Toyota Kata (Rother)