An end-user computing refresh touches every person in the organisation. Done well, it is the most visible IT improvement most users will see in years. Done badly, it is weeks of disruption and a helpdesk under siege. The difference is rarely the technology — it is the delivery model.
ClearPath Partnership delivered a large-scale end-user technology refresh for a major international airport: replacing or upgrading every end-user device with new hardware and Windows 11, while minimising disruption to critical 24/7 operations. This playbook captures the model that made it work.
Why refreshes fail
Most troubled EUC programmes share the same root causes: an incomplete picture of the estate, a deployment schedule built around IT convenience rather than user reality, manual build processes that cannot scale, and a support model that treats the day after migration as somebody else's problem. Each is avoidable with deliberate design.
Discovery: know the estate before you touch it
Every device, application and user persona needs to be understood before scheduling begins. Which applications are genuinely used? Which peripherals — scanners, printers, specialist operational equipment — must keep working on day one? Discovery is unglamorous, but it is where delivery certainty is created. The output is a segmented estate: personas and locations grouped by complexity and operational sensitivity, which becomes the skeleton of the rollout plan.
Pilot in phases, fix while it's cheap
We run structured pilots before scale deployment: first a friendly IT population, then a representative operational slice. Every issue found in pilot — a driver conflict, an application quirk, a scheduling assumption — is an issue that never reaches hundreds of users. The rollout only accelerates once pilot exit criteria are met and the per-device effort is predictable.
Automate provisioning, or drown
Manual builds do not scale past the first hundred devices. Automated device provisioning and patching — modern, cloud-driven build processes that turn a device from box to productive in a predictable time — reduced deployment effort and, just as importantly, long-term support cost: the same automation that builds a device during rollout rebuilds it in two hours when it fails in service.
Deploy around the user, not the schedule
A user-focused deployment model means users know what is happening and when, migrations are scheduled around operational reality, and each user's transition includes the small things that determine perceived success: data in place, printers mapped, favourites intact. Targeted communications — short, specific, timed to the user's own migration — outperform programme-wide broadcasts every time.
Early-life support and the BAU handshake
For a defined period after each deployment wave, dedicated early-life support sits close to the users — resolving snags directly rather than routing everything through the service desk. The programme ends with a deliberate transition into business-as-usual: documentation, known-error records and support-team training, so the operation inherits a service, not a surprise.
The playbook in brief
- Segment the estate through real discovery before any scheduling.
- Pilot in phases with explicit exit criteria; only then scale.
- Automate provisioning and patching — it pays during rollout and for years after.
- Schedule around users and operations; communicate specifically, not broadly.
- Fund early-life support and a genuine BAU handover.
Closing thoughts
“Automated provisioning was the technical linchpin here — once a device can rebuild itself from bare metal without an engineer touching it, refresh stops being a project and becomes a capability you keep.”
Gareth Wilson · Founder and Co-CEO
“Pilots aren't a formality; they're where the plan meets reality. Run them early, listen to what they tell you, and let the deployment model earn its scale — that's how you refresh thousands of devices without the operation ever noticing.”
Phil Moss · Founder and Co-CEO
Talk to the people who did the work. If the challenges in this paper look like yours, we'd be glad to share more of what we've learned — and how it could apply to your organisation.
Get in touch