Most communications modernisation programmes don't fail because the technology doesn't work. By the time a digital radio platform, dispatch console or recording system gets to procurement, the vendor shortlist has usually been through rigorous technical evaluation. The technology is rarely the problem. The programme around it is.
In two decades of leading and recovering these programmes, the failure modes repeat with remarkable consistency.
The business case assumes a clean cutover. Communications modernisation almost never happens in a vacuum — it happens while the existing system is still carrying live operational traffic, often for public safety or frontline teams who cannot tolerate a gap in coverage. Business cases that model a single cutover weekend, rather than a phased transition with fallback paths, are setting delivery up to fail before a vendor is even selected. The right question at business case stage isn't "how long will the cutover take" — it's "what happens if the cutover doesn't go as planned, and how long can we run both systems in parallel."
Governance is designed for a generic IT project, not an operational one. A standard steering committee structure works fine for a finance system upgrade. It doesn't work for a programme where field teams, dispatchers, network engineers and safety officers all need a seat at decisions that affect their operational risk. Programmes that bolt operational stakeholders onto a generic governance structure — rather than designing governance around the operational risk profile from day one — consistently lose stakeholder confidence at exactly the point they need it most: cutover.
Operational readiness is treated as a late-stage activity. Training, site readiness, subscriber planning and acceptance testing are frequently scheduled in the final quarter of a multi-year programme, as if readiness is something you do once delivery is finished rather than something you build in parallel with it. By the time readiness gaps surface, there's no schedule room left to fix them without delaying go-live.
Vendor risk is managed contractually, not operationally. A strong contract is necessary but not sufficient. Programmes that rely on penalty clauses and SLAs to manage vendor performance, without also building operational visibility into what the vendor is actually doing week to week, tend to discover problems at milestone review rather than in time to course-correct.
The disciplines that prevent this aren't complicated — they're just consistently skipped. Parallel-running cutover plans. Governance built around operational risk, not generic project structure. Readiness activity that starts on day one, not in the final quarter. Vendor oversight that's active, not just contractual. None of this is novel. What's missing is usually the discipline to apply it consistently under schedule and budget pressure, when it's tempting to assume the next milestone will go to plan.
If your programme is heading toward a cutover and any of this sounds familiar, it's worth raising now — not in the steering committee meeting after something has already gone wrong.