Map the constraint cascade before intervening: rank all steps by max capacity to predict where the next bottleneck will appear

Why This Is a Rule

Constraints don't exist in isolation — they exist in a cascade: a ranked sequence of capacity limits where removing the current constraint reveals the next one. If your 5-step workflow has maximum capacities of [100, 60, 80, 90, 70] items per week, the current constraint is step 2 (capacity 60). Fixing step 2 to capacity 100 doesn't give you 100 throughput — it gives you 70, because step 5 becomes the new constraint. Fixing step 5 gives you 80 (step 3 becomes the constraint). The cascade is: step 2 → step 5 → step 3 → step 4 → step 1.

Mapping the cascade before intervening provides two benefits: predictive planning (you know what the next constraint will be and can prepare for it) and investment sizing (if the next constraint is very close in capacity to the current one, fixing the current one produces minimal throughput gain, and you might need to fix both simultaneously to see meaningful improvement).

When This Fires

• Before investing significant effort in fixing any workflow bottleneck
• When previous bottleneck fixes produced disappointingly small throughput improvements
• When Optimize only the single slowest step — improvements to non-bottleneck steps are wasted effort regardless of their magnitude's bottleneck-only optimization needs the predictive dimension
• Complements Re-measure 2 weeks after any constraint fix: if local metric improved but total throughput didn't, the constraint has shifted elsewhere (post-intervention re-measurement) with the pre-intervention prediction

Common Failure Mode

Tunnel-vision optimization: fixing the current bottleneck without considering the cascade. After weeks of work improving step 2, throughput increases 10% instead of the expected 50% because step 5 was nearly as constrained and immediately became the new bottleneck.

The Protocol

(1) List all steps in your workflow. (2) For each step, estimate maximum throughput: how many items could this step process per period if it were the only step? (3) Rank from lowest to highest capacity. The lowest is your current constraint; the second-lowest is the next constraint in the cascade. (4) Calculate the cascade gap: if you fix the current constraint to match step 1's capacity, what throughput does the cascade produce? This is the maximum benefit of fixing the current constraint. (5) If the cascade gap is small (next constraint is close to current), consider fixing both simultaneously. If the cascade gap is large (next constraint has much higher capacity), fixing the current constraint alone will produce meaningful improvement.