A capacity model for change: how much load the organisation can carry, and where it is already spent.
Every transformation plan implicitly assumes free capacity: executive attention, middle-management change capacity, front-line absorption. In the cohort data, the median organisation begins its programme with the middle layer already committed to 118% of available capacity. The plan is drawing on an overdrawn account, and the overdraft is discovered at month nine, as slippage.
The Load Map treats change capacity the way an engineer treats structural load: rated limits per layer, current utilisation measured rather than assumed, and headroom that must exist before new load is added. It converts “can we do this?” from a judgement call into arithmetic.
Rate the limits: how much concurrent change each layer — executive, middle, front line — can absorb, benchmarked against the cohort by size and sector.
Week 1–2Measure current load from evidence: initiative registers, calendars, spans, absence data. Everything in flight is on the map, including the initiatives nobody admits to.
Week 2–4Sequence against headroom. Initiatives queue until capacity exists; the stop list is maintained as carefully as the commitment list.
Week 4+The map is re-measured at every monthly resource forum — load is a moving number, and the model is only as good as its last reading.
Portfolio planning, to size drawn phases in Draw · Hold · Release against measured rather than assumed capacity.
Programme rescue, to establish how much of a stalled portfolio the organisation can actually carry — and what must stop.
Annual planning, as the arithmetic behind the uncomfortable conversation about how much of the wish list is fundable in attention, not money.
Derived from the cohort study and applied in more than thirty-five engagements since 2024:
The Institute stewards the model and revises it after every tenth application review; the rating benchmarks are re-based annually against the cohort.
Cite as: Markham Institute, “The Transformation Load Map”, MKM-F-006, v2.0 (2026).