From Pilot to Production

From Pilot to Production

Step 1 of 9

users40accepted92%

The pilot went well: 40 users, six weeks, 92% of drafts accepted. Now everyone wants it for 4,000 people.

The Idea

A pilot answers "can this work?" Production asks "can we run it?" Before going live, check a few gates against evidence: data that refreshes on its own, a security review, monitoring that alerts someone, support that knows what to do, and an owner with a budget. Each failed gate becomes a work item, not a debate.

Real-World Example

A pop-up restaurant that sold out every night is not yet a restaurant. Before signing a lease it needs suppliers on contract, a health inspection, a manager on every shift and someone paying the bills.

The Tradeoff

Gates slow down the exciting part, and too many of them stall everything. Keep them few, objective and decided at a fixed review, then roll out in stages so the first production month can still teach you something.

Your turn

Put the steps in the right order.

1. 1Check each readiness gate against evidence↑↓
2. 2Roll out to a first wave of users↑↓
3. 3Agree the success metrics before the pilot starts↑↓
4. 4Turn each failed gate into an owned work item↑↓

Check order

Mini quiz

1 / 3

A pilot hit every accuracy target, but nobody owns the system once the pilot team leaves. The right call is:

• ALaunch now and find an owner later
• BHold the launch until an owner and a budget are named
• CCancel the project
• DExtend the pilot with no end date