After the whirlwind of Black Friday and Cyber Monday (BF/CM), the feeling is one of relief: peak demand has been met, the indicators are consolidated and operations are back to normal, at least on the surface. But it is precisely during this period, when the dust has settled and the data is fresh, that the best opportunities arise to rethink planning for 2026.
Instead of treating BF/CM just as a capacity test, it's possible to see it as a large controlled experiment, full of information about bottlenecks, slack, costs and trade-offs between service level and profitability. This is where two central concepts of mathematical modeling applied to the supply chain come in: shadow cost (shadow cost) and sensitivity analysis.
They help answer questions such as: "Where, exactly, would it be worth investing in more capacity?", "Which restriction is making the operation more expensive?" or "How risky is it to increase demand in a certain region without changing the network?".
What shadow cost and sensitivity analysis are in practice
In simple terms, shadow cost is the value associated with "loosening" a constraint in an optimization model. In a supply chain problem, a constraint can represent, for example, the maximum capacity of a DC, the number of vehicles available, the production limit in a plant or the maximum transportation budget. It indicates how much the total cost of the operation would decrease if you slightly increased that capacity.
Sensitivity analysis, on the other hand, assesses how much the optimal solution changes when you change parameters such as demand, costs, capacities, production times or deadlines. Instead of recalculating everything from scratch with every slight variation, sensitivity analysis shows which parameters are really critical and where there is room for maneuver. In a post-BF/CM context, this means using peak data to understand what really "hurt" in the operation and what was just noise.
How to use data to feed models and calculate shadow costs
Black Friday and Cyber Monday generate a density of data that doesn't exist during normal periods: picking times, volume by time window, dock occupancy, fleet utilization rate, lead time by route, returns by category, marketplace behavior versus own store, among others.
By calibrating mathematical models with this data, it is possible to reproduce the peak scenario within a supply chain optimization model (logistics network, stocks, transport flows). From there, the calculation of shadow costs provides quantitative answers to strategic questions:
- If I increased the capacity of my most critical DC by 10%, how much would the total cost of transportation and operation fall?
- If I add vehicles to a specific route during BF week, how many delays would be avoided and what would be the impact on cost per order?
- If I made the delivery window in a certain region more flexible, how much would I gain in productivity and reduced express freight?
These shadow costs help to prioritize investments: if the shadow cost associated with the capacity of a DC is high, it means that changing that constraint generates great economic benefit. If it's low, perhaps the bottleneck lies elsewhere, such as in the fleet or inventory policy.
Sensitivity analysis for different peak scenarios
The other piece of the puzzle is the sensitivity analysis, which allows you to explore scenarios such as: "what if demand in 2026 grows by 20% in a certain category?", "what if freight costs rise more than forecast?" or "what if I bring forward promotions and spread the peak over more days?". With real data from previous years, the model better reflects the dynamics of your operation, and the sensitivity analysis shows this:
- Which parameters are most dangerous to underestimate (e.g. average loading time, return rate, sorting capacity).
- Where the operation is more fragile (small variations already lead to delays, disruption or cost explosion).
- Which decisions are robust, i.e. remain sound even when demand, costs or capacity vary within a range.
In practice, this prevents the company from planning for 2026 based solely on "commercial optimism" or on historical averages that don't reflect the intensity of BF/CM. Planning is now based on possible ranges of variation, with contingency plans already built in.
From diagnosis to action
Just making a post-peak diagnosis is not enough. You have to translate shadow costs and sensitivity analyses into concrete actions. This involves reviewing the logistics network, stock strategies, logistics contracts, SLAs and even the promotional calendar. In many companies, this takes place within formal S&OP / IBP processes, which can be enriched with scenarios from mathematical models.
Instead of just discussing "if" the operation can handle the next peak, the dialogue becomes "which network design, inventory policies and transport configuration maximize margin and service level, under certain demand and cost assumptions". The BF/CM data becomes input for simulating 2026 more realistically and not just "inspiration" for retrospective presentations.
Time to turn data into strategy
The post-Black Friday and Cyber Monday period is not only the time to close reports, but the richest phase for redefining priorities and investments for the following year. With the structured use of shadow cost and sensitivity analysis, peak data stops being a hard-to-interpret historical record and becomes a clear map of where the supply chain is really constrained, where there is slack and which actions generate the greatest economic and operational return.
Linear Softwares Matemáticos works precisely at the intersection of operational data, mathematical optimization models and supply chain decision-making processes. Linear's solutions allow you to use real information from your Black Friday and Cyber Monday peak:
- Building integrated supply chain planning models, connecting purchasing, production, transfer between units and distribution.
- Running network design models that evaluate different configurations of distribution centers, stocks and modes for 2026 and beyond.
- Apply sensitivity analysis and shadow costs to identify which constraints and parameters have the greatest impact on cost, service level and capacity utilization.
- Support recurring planning cycles, in which each peak season the models are recalibrated, continually refining the strategy.
In other words, we help companies take a step beyond descriptive analysis to a prescriptive vision.
Want to find out how mathematical models can translate BF/CM learning into a more efficient and resilient operation all year round? Contact Linear and take the next step in the analytical maturity of your supply chain.
