A rollout across many sites fails in small, repeatable ways. A locked comms room. A box shipped to the wrong address. A site that went live before anyone tested it. Each one is cheap to prevent and expensive to find on the day.
This checklist walks through a rollout in the order it happens: before, during and after. It suits project managers and IT managers who deploy devices, network gear or refreshes at branches, stores, clinics or offices. Adjust the lists to your scope, but keep the order.
What should you do before a multi-site rollout starts?
Before any equipment moves, audit every site and confirm how your people will get in. Discovery fixes the scope, and access rules decide whether a planned visit can happen at all.
Do not assume the sites match. Branch offices drift over time, and the real condition rarely matches the old floor plan or the asset list.
- Build one site list with address, local contact, building type and time zone.
- Audit each site: rack or wall space, power, cabling, network uplink, current equipment and anything that blocks the install.
- Record what exists today with photos and serial numbers, so you can compare later.
- Ask the building about access: hours, loading dock, freight elevator, parking, badges and escort rules.
- Check for rules that stop work, such as quiet hours, trading hours, clinical areas or union requirements.
- Collect certificates of insurance or other paperwork the landlord needs, before the visit date.
- Name a local contact who can open doors on the day, with a backup.
- Flag unusual sites early: shared tenant buildings, secure areas, remote locations.
How do you stage and ship equipment?
Stage equipment in one place, then ship kits that are complete for each site. Configuring at a central bench is faster and safer than configuring in a closet at the site.
A kit should hold everything one engineer needs at one site, so a missing cable does not stop the visit.
- Configure, label and test devices at the staging point before shipping.
- Build one kit per site, with spares for common failures such as patch cables, power supplies and adapters.
- Print a packing list inside every box and photograph the contents before sealing.
- Label boxes with site name, contact, room and visit date.
- Record serial numbers against the site on the asset list as each kit is packed.
- Ship with tracking, and confirm delivery to the named contact before the visit.
- Plan for return shipping of old equipment, including boxes, labels and data wipe steps.
How do you choose pilot sites and schedule waves?
Pick a small set of sites that differ from each other, run them first, and schedule the rest in waves only after the pilot passes. A pilot that only covers your easiest site proves little.
Microsoft's guidance on phased Windows deployments follows the same logic. It advises recruiting pilot groups that include the broadest set of applications and devices, and separating special-purpose devices from the standard cycle.
- Choose pilot sites that cover the main variants: large and small, old and new, easy and awkward access.
- Pick pilot sites with a cooperative local contact and a lower business impact if something fails.
- Set pilot exit criteria in writing: what must pass, who decides, and what happens on a fail.
- Fix the runbook and kit list after the pilot, then freeze them for wave one.
- Group later waves by region, building type or business risk, not by convenience alone.
- Keep the first wave modest and widen later waves as the process proves stable.
- Avoid peak periods such as month end, holidays, stock takes or audits.
- Leave a buffer day between waves to fix problems found in the previous one.
What belongs in a day-of runbook?
A day-of runbook is a short, ordered document that lets an engineer finish the work without calling you. Write it for someone who has never seen the site.
- Site details: address, contact, access instructions, parking and the room to work in.
- Arrival steps: check in with the contact, confirm the change window and take before photos.
- Prerequisites: kit received, old configuration backed up, remote access ready for any remote team.
- Numbered install steps, each with an expected result.
- Rollback steps and the point at which you use them.
- Escalation path with names and numbers, and who can approve a change to scope.
- Test steps and the sign-off form to complete before leaving.
- Departure steps: remove packaging, take after photos, return old equipment and confirm with the contact.
How do you test, sign off and handle exceptions?
Test every site against the same written acceptance list, then collect a named sign-off before the engineer leaves. A site is not done when the equipment is installed. It is done when it passes the tests.
Exceptions will happen, so decide how to handle them before the first visit. Without a rule, each engineer improvises and no two sites end up the same.
- Acceptance tests: connectivity, authentication, key applications, printing, wireless coverage where relevant, and monitoring.
- Capture results as screenshots or logs, not just a tick in a box.
- Get sign-off from the local contact or a named approver on the day.
- Define what counts as an exception: blocked access, wrong or missing parts, faulty equipment, a failed test or a site that is not ready.
- Set a rule for each: retry on the day, reschedule, or escalate. Name who decides.
- Log every exception with cause, owner and new date, and review the log after each wave.
- Stop and fix when the same exception repeats. It is a process problem, not a site problem.
What goes into closeout, asset records and a hypercare week?
Closeout means every site has a report, an updated asset record and an accepted handover. Hypercare is a short period after go-live when you watch each site closely and fix issues fast.
Asset records matter because the rollout changes what you own. NIST SP 800-53 includes a control for system component inventory, and the point applies to any rollout: the inventory should match what is actually installed. Update it per site as you finish, not months later.
- Closeout report per site: work done, photos, test results, exceptions and open items.
- Asset records: serial numbers, model, location, install date, configuration version and warranty details.
- Retired equipment: removed, wiped, returned or disposed of, with proof.
- Documentation: updated site diagrams, IP plans and credentials stored in the proper system.
- Hypercare week: a named contact for each site, a simple way to report problems and a daily review of open issues.
- Check monitoring and logs for each site during the week.
- End hypercare with a short review: what broke, what to change in the next wave, and who accepts the handover.
Who owns each phase of a multi-site rollout?
Give each phase one accountable owner, even when many people contribute. Shared ownership often means no ownership.
| Phase | Owner | Key output |
|---|---|---|
| Discovery and site audit | Project manager with site contacts | Site list, audit photos, risks |
| Access and building rules | Project manager or facilities lead | Approved access for each site |
| Staging and shipping | IT or logistics lead | Tested kits and tracked shipments |
| Pilot and waves | Project manager | Pilot result and wave schedule |
| Day-of install | Field engineer lead | Completed runbook and photos |
| Test and sign-off | IT manager and site approver | Signed acceptance |
| Closeout and asset records | IT asset owner | Reports and updated inventory |
| Hypercare | Service desk lead | Issue log and handover |
Where does a field services partner fit in a rollout?
A field services partner supplies the engineers at each site, while your team keeps the plan and the decisions. It helps most when sites are spread out and your own staff cannot travel to all of them.
If you use a partner, ask for a written scope before work starts, and ask who the single point of contact is. Deploris coordinates vetted engineers from a network, with one engagement manager as your contact and a closeout report that includes photos and test results where they apply. There are no lock-in contracts.
Key takeaways
- Audit every site and settle access rules before any equipment ships.
- Stage and test centrally, and ship a complete kit to each site.
- Choose varied pilot sites, then freeze the runbook before the first wave.
- Write exception rules in advance, and log every exception.
- Finish with signed acceptance, updated asset records and a hypercare week.