The Essential Demand Planning KPIs: A Practitioner's Guide

If you've ever presented a forecast accuracy report to sales leadership only to be met with blank stares, you know the struggle. You're talking about WAPE and error distribution; they're wondering why stock-outs lost them a key account last quarter.

The disconnect often stems from which Key Performance Indicators (KPIs) we track and how we frame them.

Data is abundant in supply chain, but actionable intelligence is rare. A 60-page deck of metrics doesn't prove the value of the demand planning function. A focused dashboard that links forecast accuracy to business outcomes does.

In this guide, we'll cut through the noise. We'll cover the demand planning KPIs that actually drive performance, how to calculate them, and—crucially—how to explain them to your stakeholders.

Why KPIs Matter (Beyond Just "Accuracy")

Most demand planners obsess over forecast accuracy. While accuracy is the engine, it isn't the destination. The goal of demand planning is to enable the business to make better decisions about inventory, cash flow, and service levels.

Effective KPIs serve three distinct purposes:

1. Diagnostic: They tell you what went wrong (e.g., Bias).
2. Operational: They drive daily behavior (e.g., Exception counts).
3. Financial: They demonstrate business impact (e.g., Inventory Carrying Cost).

If your scorecard only measures how wrong you were (error rates), you're missing the chance to show how much value you added.

Forecast Accuracy KPIs

These are your diagnostic tools. They tell you the quality of the signal you are passing to the supply chain. For a deep dive into the math, check out our guide on forecast accuracy metrics.

MAPE (Mean Absolute Percentage Error)

MAPE is the most common metric, but it's dangerous for low-volume SKUs because small absolute errors on low denominators result in massive percentage errors.

• Best for: High-volume, stable items.
• Watch out for: The "division by zero" problem on intermittent demand.

WAPE (Weighted Absolute Percentage Error)

For most businesses, WAPE is the gold standard. It weights the error by volume, meaning a 50% error on your top-selling SKU hurts your score much more than a 50% error on a slow mover.

• Why use it: It aligns closely with business reality. Being wrong on the big items costs the most money.

Forecast Bias

Accuracy tells you the magnitude of the error; Bias tells you the direction. Are you consistently over-forecasting (building excess stock) or under-forecasting (risking stock-outs)?

• The practitioner's rule: A consistently high bias (e.g., +10% month over month) is often a sign of a structural issue in the model or a political issue in the consensus process.

Forecast Value Added (FVA)

FVA measures the improvement in accuracy at each step of your process. Did the sales team's override improve the statistical baseline, or did it make it worse?

• How to track: Compare Naive Forecast vs. Statistical Model and Statistical Model vs. Final Consensus.
• Why it matters: It justifies the effort. If your team spends 40 hours a week adjusting forecasts but FVA is negative, you're better off letting the "machine" run it.

Inventory KPIs

Inventory is where your forecast error goes to die. If you over-forecast, you get excess stock. If you under-forecast, you burn safety stock.

Days of Supply (DoS)

How long would your current inventory last given current demand?

• Formula: Current Inventory Value / Average Daily Usage
• The goal: Lower is generally better for cash flow, provided you don't hit service risks.

Inventory Turns

The inverse of Days of Supply. How many times a year do you cycle through your inventory?

• Context: High turns usually indicate a healthy, efficient supply chain. Low turns suggest dead stock or over-forecasting.

Stock-out Rate

The percentage of items requested that were not available. This is the direct penalty for under-forecasting (or supply disruptions).

Service Level KPIs

Service metrics measure the customer experience. You can have a perfect forecast, but if the truck breaks down, the customer doesn't care about your MAPE.

Fill Rate (Line or Unit)

• Line Fill Rate: Percentage of order lines shipped in full.
• Unit Fill Rate: Percentage of total units shipped vs. ordered.
• The trade-off: Pursuing 99% fill rates often requires exponentially more inventory than 95%.

On-Time In-Full (OTIF)

The strictest measure. Did the customer get exactly what they wanted, exactly when they wanted it?

• Why it's hard: OTIF requires perfection from demand planning, manufacturing, and logistics simultaneously.

Process KPIs

These measure the efficiency of your demand planning process.

Forecast Cycle Time

How long does it take to close the monthly cycle? If you spend three weeks gathering data and only one week analyzing it, your cycle time is killing your agility.

Consensus Attainment

Do the stakeholders actually show up and agree? Tracking attendance and sign-off rates for S&OP meetings is a valid KPI.

Plan vs. Actual Variance

This helps validate the "One Number" plan. Did the financial budget match the operational demand plan?

Financial KPIs

This is the language of the C-suite.

Revenue Forecast Accuracy

How close was the demand plan to recognized revenue? This builds trust with the CFO.

Inventory Carrying Cost

The hidden cost of holding stock (warehousing, insurance, obsolescence).

• Estimate: Often 20-30% of inventory value per year. Reducing inventory by 1M saves200k+ in straight profit.

KPI Reference Table

Here is a quick reference guide for setting up your dashboard.

| KPI | Formula | Good (Benchmark) | Great (Best-in-Class) | Frequency | |-----|---------|------------------|-----------------------|-----------| | WAPE | Σ|Actual - Forecast| / Σ Actual | < 30% | < 20% | Monthly | | Bias | Σ(Forecast - Actual) / Σ Actual | ± 5% | ± 2% | Monthly | | Fill Rate | Shipped / Ordered | > 95% | > 98% | Weekly | | Inventory Turns | COGS / Avg Inventory | 3-4x | 6-10x (Industry dependent) | Monthly | | FVA | Accuracy(Step X) - Accuracy(Step X-1) | Positive | > 5% improvement | Monthly |

Building a KPI Dashboard

Don't dump all these numbers into one spreadsheet. Tailor the view to the audience.

The Executive View

• Focus: Financials and high-level health.
• Metrics: Revenue Accuracy, Inventory Value, OTIF, Key Risks.
• Format: Trend lines and Red/Yellow/Green status indicators.

The Planner View

• Focus: Diagnostics and exceptions.
• Metrics: SKU-level Bias, WAPE by family, Top 10 deviations.
• Format: Exception lists that require immediate action.

Benchmarks by Industry

Benchmarks are tricky because "good" varies wildly by sector.

• FMCG / CPG: High volume, low volatility.
  • Target WAPE: 15-20%
• Consumer Electronics: Short lifecycles, high seasonality.
  • Target WAPE: 30-40%
• Industrial / Spare Parts: Lumpy demand, long tails.
  • Target WAPE: 45-60% (Focus on Service Level instead)

Presenting KPIs to Leadership

When you present to the S&OP executive committee, tell a story.

Don't say: "Our WAPE increased by 2 points to 24% because of a mix error in the Northeast region."

Do say: "Forecast accuracy dipped slightly, which led to a $50k increase in safety stock requirements in the Northeast. We've adjusted the baseline for next month to correct this bias."

Connect the metric to the money.

How DemandPlan Tracks KPIs

Modern software moves you away from "calculating" KPIs to "analyzing" them.

In DemandPlan, we automate the heavy lifting:

• Real-time Accuracy: WAPE and Bias are calculated instantly as actuals flow in.
• Lag Analysis: We track accuracy at different lags (Lag-1, Lag-3) to see how well you forecast the immediate future vs. the long term.
• FVA Tracking: The system automatically logs the statistical baseline, the sales override, and the final plan, giving you an automated FVA report without spreadsheets.

Conclusion

The right KPIs transform demand planning from a clerical task into a strategic function. By tracking a balanced mix of accuracy, inventory, and service metrics, you can paint a complete picture of supply chain health.

Start small. If you currently track nothing, start with WAPE and Bias. Once those are stable, layer in FVA to measure process efficiency. Remember, the goal isn't a perfect score—it's a better business decision.

---

Ready to automate your KPI reporting? See how DemandPlan can help you visualize performance and drive accountability, or explore our guide on forecast accuracy metrics for a deeper technical dive.