A Shared Certified Architect and Two Competing Deadlines

Critical path and float decide priority, not sponsor seniority.

• Project A: 14 critical-path tasks, three requiring Priya for 12 days. Losing those 12 days slips the critical path by 8 days. Project A's float is effectively zero, and the deadline is hard - a consent order with daily financial penalties. Every day of slip is a realised cost.
• Project B: Priya's tasks sit on a path with 4 weeks (20 working days) of total float and a 10-week deadline. A 12-day delay consumes 60% of that float but does not by itself threaten B's deadline.
• Conclusion: Priya's work on A has a much higher cost of delay than her work on B. The functional manager's instruction prioritises B for political reasons (the VP of Sales), not schedule reasons.

Caveat: the root problem is that the organisation has a single point of failure for a certified skill and has over-committed it 50/50. The PM cannot unilaterally reassign Priya - the functional manager owns her time - so the correct move is to make the trade-off visible with the numbers above while protecting A's critical path in the interim.

Also verify the float: if B's four weeks have already been consumed by other slippage, both projects are genuinely at risk and the escalation becomes more urgent, not less.

Before escalation:

1. Verify B's float with its PM and Priya. Confirm the four weeks is real and not already eroded. If it is gone, document that - it changes the decision.
2. Re-sequence A's critical path so Priya's three tasks sit as late as possible within the six weeks, and pull non-Priya work forward so the team is not idle while she is on B. Identify any of the 12 days that can be split into design by Priya and implementation by certified peers.
3. Meet Priya and the functional manager together with the float numbers on one page. Ask for a specific, time-boxed commitment - for example Priya on A for 6 of the 12 days in weeks 1–2, then back to B - rather than an all-or-nothing reassignment. A scheduled partial allocation often resolves this without escalation.
4. Put a currency on the exposure. Daily penalty × 8 days plus the regulatory consequence, so the escalation is an economic argument rather than a territorial one.
5. Brief your own sponsor now, not at the committee, so the decision is pre-aligned before the meeting.
6. Do not start unauthorised workarounds such as letting an uncertified person use the tooling - that trades a schedule risk for a compliance risk on a project that is already a regulatory remediation.

If the committee declines:

• Convert the slip into a managed outcome, not a surprise. Rebaseline Project A with the 8-day slip documented and put the consent-order exposure in front of the sponsor in writing. Once the committee has been told, it owns the consequence - but the PM must ensure it is told before the date, not after.
• Mitigate inside the constraint. Ask the regulator or audit body for a remediation plan with staged milestones rather than one date; regulators often accept a credible plan with interim evidence. In parallel, compress non-Priya critical-path tasks with overtime or a contractor for work that does not require certification.
• Get the decision minuted. Ask for the decision and the accepted risk to be recorded with the cost of delay attached. A decision that is not minuted is a decision nobody owns.
• Escalate through governance, not around it. Do not reassign Priya informally; that creates a second conflict and undermines the functional manager.
• Fix the structural cause. Raise the single-point-of-failure as an organisational risk and propose cross-training a second architect or contracting a certified resource next quarter, with the certification cost compared against the penalty exposure.
• Propose a priority rule. If the organisation has no method for resolving cross-project resource conflicts, recommend one based on cost of delay rather than sponsor seniority. That is the durable lesson from this case.