How the plan is built
It rests on one rule: prove it on one route or depot before you move the rest. A pilot exposes the problems that only show up with real orders, real drivers and a real day end, while the old way of working still runs everywhere else.
- Phases overlap. Training starts while set-up is still being tested. The week ranges show that.
- Each phase has an exit test, written in business terms: a day’s orders run end to end, the day end balances. Not “configuration complete”.
- The gate between pilot and rollout is a decision, taken by the people who run the operation, against the checklist below.
- Each phase has one owner by name, not a department.
Getting customers, products and prices ready to move is its own piece of work; our guide to getting distribution data ready to move sets out the sign-offs.
Implementation plan
Weeks are a starting point. Write your own dates and owners.
Project lead
Pilot route or depot
Start date
Target go-live
Phases
| Phase | Weeks | What happens | Done when | Owner | Dates |
|---|---|---|---|---|---|
| 1. Scope and data | 1 to 2 | Choose the pilot route or depot. List the processes in scope. Clean customers, products, prices and opening stock. Agree the accounting connection. | Pilot scope signed off; data loaded for testing | ||
| 2. Set-up and test | 2 to 4 | Set up routes, price lists, credit terms, devices and documents. Run real orders through order, pick, load, deliver, invoice and accounts. | One day of real orders runs end to end with no re-keying | ||
| 3. Training | 3 to 4 | Train office, warehouse and drivers on their own tasks. Drivers run a practice round on the device. | Every pilot user has done a full day of their tasks in test | ||
| 4. Pilot | 4 to 6 | Run live on the pilot route or depot. Review issues daily, and check every day end and every invoice posted to the accounts. | The go/no-go checklist below is met | ||
| 5. Fix and decide | 6 to 8 | Fix what the pilot found: settings, documents, training. Write the rollout order and the support plan. | Go/no-go agreed by the operation’s leads | ||
| 6. Roll out | 8 to 12 | Move the other routes and depots in planned waves, with the pilot team supporting each one. | Every site live; the old system read-only |
Go/no-go: before moving beyond the pilot
| Check | Yes | No |
|---|---|---|
| Orders from every channel in scope reach the warehouse without re-keying | ||
| Picks and loads match the orders | ||
| Drivers complete the round on the device: delivery proof, returns and payments | ||
| Day ends reconcile, or every variance is explained | ||
| Invoices and payments reach the accounting package correctly | ||
| Users at the next sites are trained | ||
| No blocking issue is open |
Sign-off
Operations
Finance
Date
Where implementations slip
- Dirty data found in the pilot. Duplicate customers and wrong prices surface as wrong invoices. Clean before phase 2, not during phase 4.
- Training too early or too late. Trained three weeks before the pilot, people forget; trained the day before, they panic. Train in the fortnight before.
- A pilot nobody can fail. If the go/no-go is a formality, problems move to every depot at once.
- The old process left open. Keep it as a fallback for real failures in the pilot, and watch how often it gets used. If people reach for it out of habit, the training needs another pass.
