Most supply chain leaders accept a familiar rule: you can optimise for service, cost or cash — but never all three at the same time.

Improve availability and inventory rises.
Reduce working capital and service risk increases.
Cut logistics cost and lead times stretch elsewhere.

This trade-off is treated as an operational law.

But what if it isn’t a law at all?

What if it is an artefact of how planning decisions are made?

‍

How a belief becomes a constraint

The trade-off feels real because leaders can point to years of experience where improving one dimension genuinely harmed another.

The mistake is assuming why that happened.

Traditional planning systems force decisions to be made against a single expected future:

• One demand forecast
• One lead-time assumption
• One cost scenario

When reality deviates, the impact has to land somewhere. Inventory grows. Expedites increase. Service degrades.

The trade-off appears not because the goals are incompatible, but because uncertainty has already been excluded from the decision.

‍

The average is not neutral

Planning to the average sounds sensible. In practice, it is biased.

Modern supply chains do not lose money in normal weeks. They lose money when multiple variables deviate at once — exactly the situations average-based planning ignores.

Optimisation then occurs inside an artificially narrow decision space, guaranteeing disappointment.

‍

Why optimisation keeps failing

Organisations invest heavily in:

• Forecast accuracy initiatives
• Segmentation and parameter tuning
• Governance and process maturity

Yet outcomes barely move.

Optimisation has not failed. It is being applied to a decision space that is too small.

If uncertainty is not modelled, optimisation can only produce brittle plans that break under pressure.

‍

Static plans and false control

Static plans feel reassuring because they produce clarity.

Numbers reconcile. Financials balance. Accountability is clear.

But clarity built on fragile assumptions creates surprise later.

Teams compensate informally:

• Planners override system recommendations
• Buffers are added without visibility
• Exceptions become the norm

The plan becomes a reference point rather than a decision engine.

‍

The assumption worth testing

Most planning decisions implicitly ask:

What should we do if the expected future occurs?

A more robust question is:

What decision performs acceptably across many plausible futures?

That shift reframes service, cost and cash from fixed trade-offs into quantifiable risks.

‍

What this means for leaders at this stage

At the Test Assumptions stage, the task is not to select tools or compare vendors.

It is to challenge whether the trade-offs being managed every day are truly inevitable — or simply symptoms of how planning decisions are framed.

Without that challenge, new systems will reproduce old outcomes under different labels.

‍

What this does not answer yet

This article does not describe:

• Alternative planning approaches
• Capability requirements
• How to evaluate or select solutions

‍

Sources
SoftwareVerdict, The “Goldiilocks” Zone
Uzair Bawany & JP Doggett — discussion transcripts (April 2026& November 2025)

‍