Replacing the tool that drivers use every single shift is one of the most sensitive changes a network can make. Get it wrong and you have 200 confused drivers, a flooded support line and a control room flying blind. Get it right and the new system is invisible — service runs as usual, and within days nobody remembers the old way.
This is the playbook we've seen work across networks of every size, condensed into a two-week rollout. It assumes you're moving from radio or a legacy onboard terminal to a driver mobile app (AVM/CAD-AVL). Adapt the timeline to your reality — the sequence matters more than the exact dates.
Before week one: the groundwork
The rollout succeeds or fails on preparation done before any driver touches the app. Three things have to be ready:
- A clean network plan in the system — lines, stops, schedules and calendars validated. The app is only as good as the data behind it.
- Devices decided: bring-your-own-device with a managed app, or company tablets. BYOD is faster to deploy but needs a clear support policy.
- A small pilot group identified — 5 to 10 respected drivers who will test first and become your internal champions.
Week one: pilot and prove
Start with the pilot group on real routes. The goal isn't a perfect rollout — it's to surface the 20% of edge cases that documentation never predicts: the depot with no signal, the driver who changes vehicle mid-shift, the relief that happens at a roundabout instead of the depot.
Run a daily 15-minute debrief with the pilot drivers. Fix what you can overnight, document the rest. By the end of the week, your champions should be comfortable enough to vouch for the app to their peers — which is worth more than any training slide.
Week two: scale by depot, not all at once
Roll out depot by depot, not driver by driver and not everyone on the same morning. Each depot gets a short hands-on session — 20 minutes, on their own device, logging into their own next shift. A champion from the pilot group is present at each one.
- Day 1–2: largest depot first (most edge cases, most support staff available).
- Day 3–4: remaining depots, with champions rotating to support.
- Day 5: buffer day — catch-up sessions, late starters, last questions.
The control room runs both in parallel — briefly
For the first few days at each depot, keep the old communication channel (radio) available as a safety net, but make the app the default. Pulling the safety net too early creates panic; leaving it too long means drivers never fully switch. Two to three days of overlap per depot is usually the sweet spot.
What 'done' looks like
You'll know the rollout has landed when the support questions dry up, the control room stops reaching for the radio, and the data coming back — GPS positions, schedule adherence, incident reports — is complete and trustworthy. At that point the app has stopped being a project and become the way the network runs.
The app guides a new driver through every stop. Onboarding dropped from weeks to a few hours.
Two weeks is realistic for a 200-driver network when the data is clean and the rollout is sequenced. The single biggest predictor of success isn't the technology — it's whether your pilot champions believe in it before their peers ever log in.