Your Supply Chain Is Probably Lying to You

The most dangerous KPI isn't the one that's wrong. It's the one that's technically correct but tells you the wrong story.

We all love KPIs.

On-time delivery: 96%.
Forecast accuracy: 91%.
Inventory: down 12%.
Manufacturing efficiency: 94%.

Sounds like we're doing a pretty good job, right?

Well… maybe.

One of the things I've learned working with supply chains is that a KPI can be perfectly correct and still give you the wrong answer.

Imagine a company with a 95% On-Time Delivery rate.

At first glance, that's great.

But then you ask a few more questions:

- On time according to the original customer request date, or the date we changed it to?

- How many orders were delivered on time because the customer accepted a later date?

- How many required an expensive expedited shipment to make the deadline?

- And how many orders were already late before someone realized there was a problem?

Suddenly, that 95% doesn't look quite as impressive.

The same happens with lead time.

Suppose we reduce our average end-to-end lead time from 30 days to 25.

Success!

But what if the average improved because a few large orders moved much faster, while smaller orders became significantly more unpredictable?

The average tells us we improved.

The customer experience might tell us something completely different.

And that's the problem with averages and isolated KPIs.

They measure what happened. They don't necessarily explain why it happened.

This is why I think the next step in supply chain performance management isn't simply adding more KPIs.

It's connecting the KPIs and understanding the story behind them.

Instead of asking “What's our lead time?” we should ask “Where is the time actually being spent?”, is it manufacturing? Transportation? Waiting for an approval? Quality release? Planning? Customs? Or simply waiting for someone to make a decision?

And instead of asking “Are we on time?” maybe we should ask “What did it take to be on time?”

Because there's a huge difference between delivering on time through a stable, predictable process and delivering on time after three escalations, an emergency production slot and an expensive air shipment.

Same KPI. Completely different supply chain.

That's why I believe good supply chain analytics should work a bit like detective work.

The KPI tells you where to look, the process tells you what happened and root-cause analysis tells you what needs to change.

So the next time your dashboard is full of green numbers, don't just celebrate.

Ask a few uncomfortable questions, what is this KPI really telling us? What is it hiding? and perhaps most importantly: Are we measuring performance… or just measuring the symptoms?

Because sometimes the biggest supply chain problem isn't a bad KPI, it's a KPI that makes us believe everything is fine.

Antifragility: The Next Evolution of Supply Chain Design

For decades, supply chains have been optimized around efficiency.

After repeated shocks, from natural disasters to trade wars, to COVID-19, the focus shifted to resilience. Today, organizations are exploring a more radical idea: antifragility.

Coined by Nassim Nicholas Taleb, antifragility describes systems that benefit from disorder. While fragile systems break under stress and resilient systems merely survive it, antifragile systems improve because of volatility. Applied to supply chains, this concept challenges many long-standing assumptions about cost, control, and risk.

A resilient supply chain is designed to absorb shocks and return to its original state. It relies on buffers, backup suppliers, safety stock, and recovery plans. These are necessary and valuable but they still assume that disruption is an exception.

Antifragile supply chains start from different premises: disruption is normal.

Rather than asking, “How do we recover faster?”, antifragile systems ask: “How do we learn faster and become stronger each time disruption occurs?”

This subtle shift has profound implications.

Antifragility does not mean chaos, nor does it mean abandoning efficiency altogether. Instead, it involves intentional design choices that allow learning, adaptation, and optionality.

Key characteristics include:

1. Redundancy with purpose

Traditional supply chains view redundancy as waste. Antifragile networks treat redundancy as a strategic asset, multiple suppliers, routes, and production options that can be tested under stress to reveal which perform best.

2. Small, frequent stressors

Rather than avoiding failure at all costs, antifragile organizations allow small failures, missed forecasts, supplier switches, pilot disruptions, to surface weaknesses early, before catastrophic breakdowns occur.

3. Optionality over optimization

Instead of committing fully to the lowest-cost supplier or single global network, companies maintain options: regional sourcing, flexible contracts, modular production, and postponement strategies.

Despite its appeal, antifragility remains uncommon in supply chain design.

One reason is measurement. Traditional KPIs; cost, service level, asset utilization etc reward stability and penalize redundancy. Antifragility delivers value over time, often invisibly, until a major shock reveals its advantage.

Another barrier is culture. Antifragility requires leaders to tolerate controlled inefficiency, empower local decision-making, and accept that not all failures should be eliminated only the catastrophic ones.

Finally, antifragility challenges the legacy of lean thinking. While lean principles remain powerful, applying them without regard for volatility can unintentionally increase fragility at the network level.

Global supply chains are unlikely to become calmer. Climate risk, geopolitical fragmentation, regulatory divergence, and technological disruption are increasing variability, not reducing it.

In this environment, the question is no longer whether supply chains should be resilient but whether they should be designed to evolve through stress.

The most competitive supply chains of the future may not be the most efficient ones but the ones that grow stronger every time the world changes.

Balancing OEE with Other Supply Chain KPIs: Navigating the Tradeoffs

In the world of operations, few metrics get as much attention as Overall Equipment Effectiveness (OEE). OEE measures how effectively a manufacturing asset is utilized by combining three factors: availability (uptime), performance (speed vs. ideal cycle time), and quality (good units produced vs. total units).

In simple terms, it’s a snapshot of how close a machine or line is to running at its theoretical maximum potential. Improving OEE is often seen as a direct path to better productivity and lower costs, but like many metrics, focusing on it in isolation can create conflicts with other critical supply chain goals.

For instance, pushing OEE higher often means striving for longer production runs and fewer changeovers. That’s good for machine efficiency, but it can hurt inventory turns and customer responsiveness. A plant that maximizes OEE by producing large batches of a single SKU may end up tying up working capital in excess stock and reducing the ability to adapt to shifting demand. Similarly, prioritizing OEE can clash with on-time delivery if equipment schedules are optimized for efficiency rather than customer requirements.

Another tradeoff emerges with flexibility and innovation. To keep OEE high, operations teams may resist frequent product launches or engineering changes, both of which introduce downtime, lower yields, and slower cycle times. Yet in today’s market, agility and product variety often matter just as much as asset utilization.

So how do you balance these competing priorities? The key is to treat OEE not as an end in itself, but as one piece of a broader performance puzzle. A mature operations strategy aligns OEE with business objectives by:

• Defining the right horizon: Short-term dips in OEE may be acceptable if they support long-term goals like faster customer response or product diversification.
• Using tiered KPIs: Pair OEE with customer-facing measures such as fill rate, lead time, and service level, ensuring that equipment efficiency doesn’t come at the expense of market performance.
• Driving continuous improvement, not perfection: The pursuit of 100% OEE is unrealistic. Instead, focus on targeted improvements that also strengthen supply chain resilience.

In the end, OEE is a powerful tool for uncovering hidden losses and improving operations, but it should never overshadow the broader mission: delivering the right product, at the right time, at the right cost.

Balancing OEE with other KPIs ensures that efficiency gains translate into true supply chain value.

Forecast bias

In this post we are going to touch on one of the most important KPIs that any demand planner should focus on: Forecast bias

Forecast bias is the general tendency for a forecast to be higher or lower than the actual value.

Forecast bias is distinct from forecast error in that a forecast can have any level of error and still be completely unbiased. For instance, even if a forecast is fifteen percent higher than the actual value half of the time and fifteen percent lower than the actual value the other half of the time it has no bias. If the forecast is on average fifteen percent higher than the actual value has both fifteen percent error and fifteen percent bias.

Bias can exist in statistical forecasting or in judgment methods. With statistical methods, the forecasting model must be adjusted or switched to a different model. For judgment methods however, bias can be conscious and driven by certain incentives provided to the forecaster or it can be unconscious.

How to simply calculate forecast bias at an aggregated level?

BIAS = Historical Forecast Units minus Actual Demand Units.

The impact of bias can mean that either an organization is holding too much inventory (over-forecast bias) or missing sales due to service issues (under-forecast bias).