Turn an idea into a controlled experiment and a useful, supportable first offer.
Your learning outcomes
Distinguish technical, commercial and operational readiness
Design a minimum evidence-generating prototype
Write a pilot with acceptance criteria and stop rules
CLASS 4.1 / 60 MINUTES
Readiness is more than a working demonstration
Define the claim before the test
A demonstration shows something happened under particular conditions. A product claim says what a customer can reasonably expect in its own context. Write the claim, intended user, environment and limits before designing a test. SignalBox could initially claim only that it presents a vibration trend for a maintenance review, not that it predicts every failure. The distinction changes the required evidence and the consequences of error. Keep speculative future capabilities separate from the behaviour actually tested in the current version.
Map several kinds of readiness
Assess the technology, customer proposition, rights position, manufacturing or deployment process, support model and team capability separately. A laboratory result can be technically promising while installation, data access or commercial demand remains uncertain. Use a simple red, amber and green map with evidence beside each rating. Do not treat a single numerical readiness label as proof that the whole business is ready. The map should expose the weakest dependency that blocks the next decision, helping the team spend effort where it changes the outcome.
Identify risk before increasing exposure
Ask who could be harmed by a wrong result, an outage, a security failure or misuse. Consider both ordinary errors and plausible misuse. Set boundaries around intended use and define when a person must review a result. In the running case, a monitoring experiment should not silently replace safety-critical maintenance procedures. Consult the relevant specialists before tests affect real operations. A careful team can learn from a narrow, controlled trial without making broad claims or exposing customers to consequences it has not planned to handle.
PUT IT TO WORK
Your activity
Write a bounded product claim and a readiness map for the first version. Identify one technical, one commercial and one operational uncertainty.
Produce this
A product claim, readiness register and safety boundary.
CLASS 4.2 / 60 MINUTES
Prototype to answer the expensive question
Choose the uncertainty, then the prototype
A prototype is a learning instrument. Decide which uncertainty is most important before choosing the interface or technology. A paper workflow may test whether a report supports a decision; a bench rig may test sensing performance; a manual concierge service may test the value of an outcome. Building a polished dashboard is wasteful if the unresolved issue is whether anyone acts on the information. Choose the smallest ethical experiment that can change a decision, and be honest about any manual work hidden behind the demonstration.
Use a relevant baseline
Compare the prototype against the current process or another credible alternative. Define the task, test conditions and success measure before seeing the results. Record settings and failures so the experiment can be repeated. A faster response is not automatically better if it increases false alarms or creates extra work. For SignalBox, compare the proposed report with the information a maintenance lead ordinarily receives. Use representative cases where possible and state how laboratory or synthetic data differs from the real deployment environment.
Turn results into design decisions
After a test, distinguish defects, unmet requirements and assumptions that were wrong. A defect may need a repair; a wrong problem assumption may require changing direction. Do not count a feature as validated merely because the team managed to demonstrate it. Explain what was learned, which limitations remain and what must happen before broader use. Archive the tested version and protocol. A useful prototype report can recommend removing a feature, simplifying a workflow or ending an approach, rather than always adding another development phase.
PUT IT TO WORK
Your activity
Design a low-cost prototype experiment with a baseline, success criterion, failure cases and a version record. Use synthetic data where real data is unavailable.
Produce this
A repeatable experiment protocol and a decision template.
CLASS 4.3 / 60 MINUTES
Run a pilot that can lead to a decision
Agree scope and responsibilities
A pilot needs a clear purpose, limited duration, named participants and a shared definition of success. Agree who installs, trains, supports, approves access and handles incidents. Identify data permissions and any operational restrictions before starting. Explain whether the pilot is paid, free or discounted without treating the arrangement as evidence of normal pricing. A friendly trial can still consume substantial customer time. Respect that cost by specifying the question it answers and what both parties will receive at the end.
Set acceptance, withdrawal and rollback
Define measurable acceptance criteria linked to the intended use. Include reliability, usability, support effort and negative outcomes, not only the most flattering metric. State the conditions that stop the pilot and how ordinary operations will continue if the product is removed. For SignalBox, a pilot might test report usefulness without allowing automated machine shutdown. Prepare a communication route for problems and preserve the customer's ability to withdraw. A pilot agreement and any legal terms should be reviewed appropriately rather than copied from this lesson.
Close the loop commercially
At the review, compare results with the agreed criteria and record disagreements. Calculate the effort required to deliver the trial, including exceptional manual support. Ask what would have to change for a normal purchase, who approves it and what procurement evidence remains missing. Do not present a successful pilot as a guaranteed recurring contract. Produce a short report with the recommendation, limits, costs and next decision. The outcome may be a paid deployment, a redesigned offer, another bounded test or a decision not to proceed.
PUT IT TO WORK
Your activity
Write a four-week illustrative pilot plan, including acceptance, incident handling, withdrawal and a commercial review. All timing is a teaching assumption.
Produce this
A pilot charter and acceptance checklist.
MODULE ASSIGNMENT / SELF-REVIEW
Build the evidence.
Prepare a minimum-evidence product plan with a readiness map, prototype protocol and pilot charter. Explain how the experiment could disprove the commercial proposition.
Assignment self-review rubric
Criterion
Points
Bounded claim and risk awareness
25
Relevant baseline and repeatable test
30
Pilot responsibilities and acceptance
30
Commercial learning and stop conditions
15
Use the rubric to evaluate your own work or ask a suitable mentor. Written assignments are not uploaded or automatically marked.