Launch dates create focus. They also create a misleading finish line. A product can ship on time, attract initial attention and still fail to become useful. The more revealing review happens after novelty has faded and operating reality has accumulated.

The 18-month test asks whether a technology investment is adopted, dependable, economically viable, owned by the operation and important enough that users would notice if it disappeared.

This test does not require waiting eighteen months to manage. It changes the questions asked from the beginning.

Adoption beyond registration

Measure the intended users completing the intended workflow repeatedly. Accounts, licences and first logins are distribution metrics.

Segment by role, customer and use case. Look at retention of use, successful completion and workarounds. Interview people who stopped. They often reveal missing fit, trust or integration.

An internal mandate can create usage without value. Pair adoption with outcome and satisfaction.

Reliability where consequences live

Review uptime, latency, defects, support volume, data quality and recovery. Focus on the business workflow: a service may be technically available while invoices fail or users cannot complete checkout.

Include dependency and vendor incidents. Test recovery and restore, not only the existence of a backup.

Track whether operational burden is increasing. A product that survives through heroic manual support has not achieved durable reliability.

Economics after scale

Compare total operating cost with the business value or capability protected. Include licences, cloud, model use, support, engineering, compliance and change.

Measure unit cost as volume grows. Confirm forecast savings were converted into capacity, margin, revenue or reduced risk. Replace business-case assumptions with observed evidence.

Avoid judging enabling platforms solely by direct revenue. Their outcome may be release speed, resilience or risk reduction, but that outcome still needs an owner and measure.

Customer or employee dependence

Ask what people would do if the technology disappeared tomorrow. Would a meaningful workflow stop, slow or become less accurate? Have teams built trusted routines around it?

Dependence should not come from lock-in or lack of export. Healthy dependence means the product performs a valuable job well.

Review complaints and requests as evidence of engagement, not only dissatisfaction. Users who depend on a system often care deeply about its details.

Operating ownership

Identify who owns roadmap, quality, cost, security, data and adoption. Confirm knowledge is distributed, documentation is current and change decisions are made through a visible process.

If the original project team left and nobody owns improvement, the investment is decaying. If a vendor is essential, confirm continuity, portability and exit plans.

Learning and strategic fit

Compare the problem now with the problem at approval. Markets and operations change. A responsible review can conclude that a well-delivered product no longer deserves further investment.

Record what the organisation learned about users, technology and delivery. Option value can matter even when an experiment stops, provided the learning is specific and informs a decision.

Vinove’s model is to build, own and stay. The standard behind its five companies is not how impressive technology looks at announcement, but whether it continues to work where customers depend on it.

A review scorecard

At 3, 6, 12 and 18 months, review adoption, user outcome, reliability, security, unit economics, support burden, operating ownership and strategic relevance. Classify the investment: scale, improve, maintain, reposition or retire.

The 18-month test is a discipline against launch theatre. It makes durability part of the original design and gives leaders permission to invest, change or stop based on how technology performs in the world that actually uses it.

Design the test at approval time

The investment case should name the 3-, 6-, 12- and 18-month evidence before delivery begins. At three months, the question may be successful adoption by a defined group. Six months may test quality and support. Twelve months can examine unit economics and strategic fit. Eighteen months asks whether the capability remains depended on and responsibly owned.

Assign data sources and owners for each measure. Capture the baseline now; reconstructing it later creates weak comparisons. Include a retirement condition so stopping is part of responsible ownership.

This design also improves launch decisions. Teams know that a feature count will not be enough, so they invest earlier in adoption, instrumentation, documentation and operating responsibility. The long view becomes a product requirement rather than a retrospective philosophy. The Vinove Standard provides a related decision test for evidence, ownership and endurance.