The reorder point on the chocolate biscuit ingredient was set in February. It made sense in February. Daily consumption was steady, the supplier was reliable, and a minimum threshold of eight hundred kilograms gave the team about ten days of cover. Then November arrived. Christmas demand pulled production volumes up by three times, daily consumption jumped from seventy kilograms to two hundred and ten, and the reorder point that gave ten days of cover in February now gave three. By the time the alert fired, the line was already dry. Procurement scrambled, paid premiums, and the post-mortem concluded what every post-mortem in this category concludes: someone should have updated the threshold.
The problem with that conclusion is that it puts the burden on a human to remember to retune a static number every time demand shifts. In a fast-moving consumer goods operation with hundreds of stock-keeping units across multiple production lines and several seasonal cycles per year, no human is going to keep up. The reorder points drift out of sync with reality, and the system starts generating signals that are either too early (tying up working capital) or too late (causing stockouts). Either way, the cost is real.
The fix is not better discipline around updating thresholds. The fix is a different model for how reorder points get calculated in the first place. Seasonal reorder point calculation belongs in the planning logic, not in the static configuration of an item record.
Why Static Reorder Points Fail in FMCG
A static reorder point is a fixed quantity that triggers replenishment when stock falls below it. The number is usually derived from average daily consumption multiplied by lead time, plus a safety buffer. The math is sound. The problem is that every variable in the equation moves over the course of a year, and a static number cannot move with them.
In FMCG, the variable that moves the most is consumption. Demand seasonality FMCG businesses face is rarely subtle. Ice cream peaks in summer. Hot chocolate peaks in winter. Sun cream, school supplies, and barbecue charcoal all have curves where the peak month consumes ten times what the trough month consumes. A reorder point set at average consumption is wrong every single day of the year. It is too high in the trough (causing overstock) and too low at the peak (causing stockouts). The only time it is approximately right is during the brief transition windows on either side of the curve.
The second variable is lead time, which often shifts seasonally as well. Ingredient suppliers that quote a five-day lead time in the slow season may take ten days in the peak, because every customer in the category is ordering at the same time. A reorder point built on five-day lead time becomes structurally inadequate when the realised lead time doubles. Our piece on reactive to predictive procurement goes deeper into how lead time data gets stale, but the shorthand is that suppliers slow down exactly when you most need them to be fast.
The third variable is the safety buffer itself. A buffer sized for the volatility of a normal month is undersized for the volatility of a peak month. Demand is not just higher during the peak. It is also more variable, more sensitive to promotions, and more prone to spikes that no monthly average can capture. Static reorder points do not just fail at the average. They fail at the variance.
Reading the Consumption Signal Properly
The starting point for any seasonal reorder logic is understanding what your consumption data is actually telling you. Most teams look at average daily usage over the past thirty days. This is the wrong window for seasonal goods. A thirty-day average lags reality by at least two weeks, and during a transition into peak season, two weeks is the difference between covered and stocked out.
A better signal is rolling consumption over multiple windows of different lengths, weighted toward the most recent data. A seven-day window tells you what is happening right now. A fourteen-day window tells you the short-term trend. A thirty-day window provides context. When the seven-day window is materially higher than the thirty-day window, demand is accelerating, and reorder logic should respond accordingly. When the seven-day is lower, demand is decelerating, and you can afford to pull back on order quantities.
FalOrb tracks daily consumption rates per item with trend indicators and anomaly detection. When the consumption rate moves more than two standard deviations from the established baseline, the system flags the change. This is not a generic stock alert. It is a specific signal that the underlying demand pattern has shifted, and any reorder logic that depends on the old baseline is now suspect. The planner can investigate, confirm whether the shift is a genuine seasonal acceleration or a one-off spike, and adjust the planning posture accordingly.
The point of consumption pattern analysis is not to predict the future perfectly. The point is to detect the change quickly enough that the planning system can respond before the old assumptions cause a stockout. In a seasonal business, the old assumptions are wrong by definition for at least a third of the year.
Letting Horizons Govern Replenishment
Once the consumption signal is being read properly, the next question is how to translate it into action. This is where horizon-based planning earns its keep. Instead of relying on a single reorder point that has to be retuned constantly, you let the MRP engine recalculate net requirements across multiple planning windows and surface the gaps wherever they appear.
A seasonal MRP setup runs against horizons of seven, fourteen, thirty, and sixty days, with each horizon answering a different question. The seven-day view checks immediate cover against current consumption. The fourteen-day view checks whether incoming purchase orders will arrive in time to support continued consumption at the current rate. The thirty-day view extrapolates the consumption trend forward and identifies whether seasonal acceleration will outpace the planned receipt schedule. The sixty-day view sets the strategic posture: does the supplier need a heads-up that volumes will be three times normal next quarter?
The key shift is that the horizon, not the static threshold, decides whether stock is sufficient. A material with eight hundred kilograms on hand might be sufficient in March (when consumption is seventy kilograms a day) and critically short in November (when consumption is two hundred and ten). The same physical inventory level produces different planning outcomes because the consumption rate going into the projection is different. The system does the math automatically. The planner does not have to remember to flip a switch.
This approach is explored further in our deeper treatment of MRP planning horizons. The relevant point for seasonal goods is that horizons make the planning logic consumption-aware in a way that static reorder points cannot match. The reorder point becomes an emergent property of current consumption, lead time, and the production schedule, rather than a number entered six months ago and forgotten.
Dynamic Safety Stock for Seasonal Volatility
Safety stock deserves its own treatment because it is the variable most often mishandled in seasonal planning. The traditional approach is to set a single safety stock value per item and leave it. The result is the same problem as static reorder points: the buffer is wrong most of the year and only right during transitions.
Dynamic safety stock acknowledges that the appropriate buffer is a function of two things that change seasonally: the variability of demand and the variability of supply. During the peak, demand variability rises because promotional spikes, weather effects, and downstream customer ordering patterns all become more volatile. Supply variability often rises in parallel because suppliers are stretched. The safety stock buffer should expand to reflect both effects, then contract again when the peak passes.
In a system designed for this, safety stock is not a static field on an item record. It is a calculated value that responds to recent consumption variance and recent lead time variance. When consumption becomes more volatile, the buffer grows. When lead times stretch, the buffer grows. When both calm down, the buffer shrinks, freeing up the working capital that was tied up in inventory you no longer need to hold.
FalOrb's MRP engine recalculates net requirements after every stock movement, after every production order confirmation, and on a scheduled background cycle. When the underlying consumption pattern shifts, the recalculation picks it up automatically. Alerts that fired when the old pattern was in effect auto-resolve as conditions change, and new alerts surface when the new pattern creates new gaps. The planner is not chasing a static number. The system is continuously reconciling current reality against the production schedule and surfacing the deltas as they appear.
Closing the Loop With Auto-Resolution
The final element of seasonal reorder logic that often gets overlooked is what happens when the season ends. A reorder point that was tuned for peak demand becomes wildly oversized when demand returns to baseline. If no one remembers to retune it, the system starts generating reorder recommendations that sit far above what is actually needed, working capital balloons, and the warehouse fills with material that will not be consumed for another nine months.
Auto-resolution is the structural answer. When the consumption rate drops back to baseline and projected available balance comfortably exceeds projected demand across all planning horizons, the alerts that drove peak-season ordering should resolve themselves. The reorder recommendations should disappear. The system should reflect the post-peak reality without anyone having to clean up the configuration manually. FalOrb auto-resolves alerts when the underlying condition clears, and the restock intelligence engine auto-dismisses recommendations when the shortfall they were addressing no longer exists.
This is what makes a seasonal MRP setup actually maintainable. The planner is not doing semi-annual reconfiguration projects. The system is responding to the consumption signal continuously, expanding its planning posture when demand grows, and contracting it when demand fades. The reorder point, in this model, is not a number you set. It is an outcome of how the system interprets current consumption against the production schedule and the supplier lead times in play that week.
The teams that get this right stop talking about reorder points at all. They talk about horizons, consumption trends, and supplier postures. The numbers underneath are still there, but they are derived rather than configured. And the chocolate biscuit line never runs dry in November because the system caught the consumption acceleration in September.
FalOrb helps manufacturers replace static reorder points with horizon-based planning that adapts to seasonal demand patterns automatically. Book a 30-minute walkthrough or email us at [email protected] to see how it applies to your operation.