03. The Best Demand KPI? Simple Is Simply Better
Statistical metrics like MAD and MAPE have their place in demand planning, but I've noticed something interesting: the moment these terms enter a Demand Review discussion, attention starts to wander by at least half the group. This observation reinforces to me one of the most effective - and simplest - KPIs I've encountered in demand planning - one that's often overlooked precisely because of its simplicity: the hit/miss measure.
A simple binary measure in this world of sophisticated analytics? Yep.
The concept is straightforward. You set a tolerance band (let's say +/-10%), and every SKU either hits within that band or misses it. If you forecast 100 units and sell anywhere between 90 and 110, that's a hit. Outside that range? It's a miss.
Add up the number of hits and divide by the total number of SKUs. That's it. No complex calculations, no weighted averages, no statistical gymnastics.
Here's why I like this measure:
Universal Understanding: When you tell your team "65% of our SKUs were within tolerance last month," everyone gets it. From the newest demand planner to the Sales Director, there's no room for misinterpretation. Present it with simple traffic light visuals (red for miss, green for hit), and you've got instant visibility. Throw in a trend line from the last 6 months for good measure.
Actionable Insights: The simplicity drives better conversations. When 35 of your SKUs miss the target, and 28 of those were over-forecasted, you've also got a signal of optimism bias in your planning process. There's an immediate takeaway for you to address in upcoming forecast rounds. Forecast accuracy and forecast bias in one 😊
Clear Progress Tracking: Watching your hit percentage climb from 60% to 70% to 80% is both motivating and unambiguous. It's hard to argue with improvement when it's that clear.
The good news and, ironically, the bad news? This KPI is simple to understand and effective at driving engagement.
Unfortunately, the simplicity of this measure can work against it. In today's world of machine learning and AI, suggesting a simple hit/miss binary measure can feel like bringing a calculator to a quantum computing convention.
But what I've seen after years in the field is this basic measure drives more meaningful improvements than many complex metrics, especially in organisations just starting their IBP journey where simple is good. It's particularly effective at driving cross-functional alignment - when everyone from Supply Chain to Sales understands the metric, better conversations happen.
Should we abandon MAPE, MAD, and other statistical measures? Of course not. They absolutely have their place, especially as an organisation's forecasting maturity grows. But if you're struggling to get traction with forecast accuracy improvements, or if your team's enthusiasm wanes every time someone mentions weighted absolute percentage error, maybe it's time to give the humble hit/miss measure a try. Ultimately, good demand planning isn't about having the most sophisticated metrics - it's about driving better discussion, decisions and actions.
Start with the basics, get them right, then do the fancy stuff.
What's your experience with forecast accuracy measures? Have you tried the hit/miss approach? Share your thoughts - your experience might be exactly what another planner needs to hear.
#IBP #DemandPlanning #Forecasting #SupplyChain
About me: I'm just a regular IBP guy who's seen enough forecast accuracy measures to know what works and what doesn't. And yes, I've overcomplicated things myself plenty of times before learning to embrace simplicity.
Want to learn more? Visit me at www.planninglab.co.nz