Products must be tested at the duration and intensity of their real use

Products must be tested at the duration and intensity of their real use

A product that works in a short demonstration can still fail when a customer depends on it. The evidence becomes meaningful when the test resembles the real duration, intensity, variation, intervention, failure, recovery, and support conditions of use.

The point of dependence defines the test

The right duration is not automatically long. It is the period over which the product has to deliver the claimed value.

For a transaction system, the useful unit might be a peak traffic window. For an endurance product, it might be several hours of motion, heat, load, and fatigue. For a robot, it might be a work shift that includes setup, ordinary operation, interruption, maintenance, and recovery.

The test should reach the moment when the user has no easy substitute and the cost of failure becomes real.

Intensity is broader than physical force

Intensity may mean speed, concurrency, precision, cognitive demand, transaction volume, object variation, network pressure, or consequence. A product can pass at low intensity and fail when the actual workload changes its state.

That does not make a low-intensity test useless. It makes the claim narrower. Early tests can validate a component, while a readiness decision needs the intended load.

Intervention is part of the product

Setup, monitoring, approvals, resets, repairs, and exception handling should appear in the result. A human intervention can be a valid operating control. Hiding it creates a false impression of autonomy and cost.

The same rule applies to support. A system that works with its builders present may be a strong prototype. It is not yet the same product that an ordinary operator can understand, maintain, and recover.

Failure and recovery should be specified in advance

A test needs a success measure and a failure threshold before the run begins. Otherwise, the team can reinterpret an incomplete or unsafe result after seeing it.

The record should capture the unsafe or degraded state, how it was detected, whether a person could intervene, which safeguard activated, how service resumed, and whether the same condition can recur.

E008 and E119 show the same principle at different scales

E008 applies the principle to Blackwell, Figure 01, and SIMA. A platform launch, research report, or selected robot sequence provides real evidence, but it does not reproduce the duration and intensity of deployment.

E119 applies the same idea to endurance products. Short use did not reveal the pressure, fit, heat, movement, and access problems that appeared during longer activity.

The shared thesis is not that every product needs one enormous test. It is that the test has to resemble the use being claimed.

Boundaries

This note is a decision lens. It does not replace regulated testing, safety engineering, clinical validation, security assessment, accessibility review, certification, or a domain-specific protocol.

Current performance claims, standards, product states, and named test results require fresh primary-source review. The core Venture Step sources are [[Products Must Be Tested at Real Duration and Intensity]], [[E008 Article]], [[E119 Article]], [[E119 Full Transcript]], and [[E119 Evergreen Ideas]].