Resource Management in Multi-Project Environments
The assumption embedded in most project management literature is that the project manager has a dedicated team. The reality for most practitioners is quite different. Your lead developer is also supporting three other projects. Your subject matter expert has a day job they have never left. Your business analyst splits time between your project and a BAU backlog. This is the shared-resource environment — and it is not an exceptional situation to be managed around, but the normal operating condition for project management in most organisations. The question is not how to avoid it but how to make it work.
The real costs of shared resources
Resource contention. When two projects need the same person at the same time, one project waits. That wait is rarely planned — it surfaces as a schedule slip that was not on anyone's risk register because the conflict was not visible until it happened. At the portfolio level, resource contention is the single most common driver of schedule variance that project managers describe as "external."
Context-switching cost. Research on cognitive task-switching consistently shows that moving between different types of work carries a switching cost. Gerald Weinberg's early work on software development estimated that a person working on two projects simultaneously loses approximately twenty percent of their productive capacity to switching; three concurrent projects loses approximately forty percent. These are estimates, not universal constants, but they reflect a real phenomenon: the time to re-establish context, recall where you left off, and rebuild mental models of the work is not trivial, and it accumulates across every transition in a working day.
Unresolved priority conflicts. When a shared resource receives conflicting demands from two project managers with equal authority, they typically resolve it themselves — based on their own assessment of urgency, their relationship with each PM, or simply who asked most recently. This is not a resourcing problem; it is a governance problem. The absence of a clear priority decision pushes the decision down to the person least equipped to make it.
Unplanned absences and BAU spikes. Shared resources are usually shared because they have a primary role outside the project. When that primary role demands unexpected attention — a production incident, a quarter-end crunch, an urgent customer issue — the project is the first thing to lose the resource. This is structurally different from planned leave and harder to buffer.
Portfolio-level resource visibility
The most important intervention is making resource demand visible at the portfolio level. This does not require sophisticated software. A simple capacity plan — a spreadsheet that maps each shared resource against all active projects, by week or by month, showing allocated percentage and available capacity — reveals conflicts that are invisible when each PM manages their project in isolation. Maintaining this at the PMO or portfolio level (rather than project level) is essential: a PM can see their own demand on a resource but cannot see competing demands from other projects unless someone aggregates them. The capacity plan becomes the factual basis for resourcing conversations that would otherwise be political.
Explicit prioritisation
When conflicts arise — and they will — the resolution must come from an authority above the project level. The project manager's role in a conflict is to surface it promptly, present the options and their consequences clearly, and escalate to the portfolio sponsor or governance body for a decision. The mistake is either to absorb the conflict quietly (hoping it resolves itself) or to compete with the other PM at the resource level. Neither approach produces the right answer. Explicit priority ranking of projects at the portfolio level — agreed in advance, not decided in the moment of conflict — gives both the resource and the PMs a clear basis for decisions without requiring sponsor intervention every time.
Dedicated teams and buffer management
Where the schedule consequence of context switching is high enough to justify it, advocate for dedicated assignment. A resource assigned to a single project for its duration — even if the project only runs for six weeks — delivers significantly more effective throughput than the same resource at fifty percent across two simultaneous projects. When dedicated assignment is not possible, build explicit buffer into the schedule for the transitions you can predict. Do not plan for a shared resource at one hundred percent capacity when their actual available capacity, accounting for BAU and switching, is closer to sixty. The gap between planned and real capacity is where most schedule slippage is generated.
If your organisation is managing a complex portfolio with shared resources and finding that project schedules consistently slip for the same reasons, XNM's program and project delivery practice can help you design a portfolio-level resource management approach that makes conflicts visible before they become schedule problems.