The Thursday Problem
Retail demand is not evenly distributed across the week. For most brick-and-mortar formats, Friday through Sunday represents 50–60% of weekly unit volume. This isn't a surprise to anyone who runs physical retail. What is a surprise — or at least a consistent gap in planning execution — is how often brands fail to translate that knowledge into their DC-to-store allocation timing.
The allocation window that matters is Thursday. Stock that arrives at a store location by Thursday morning is available for the weekend selling peak. Stock that arrives on Friday afternoon is partially available. Stock that arrives on Monday captures none of the weekend, and depending on what sold out Friday and Saturday, may arrive to find the promotional context has passed.
This sounds obvious when stated directly. But the allocation decisions that determine what ships on Thursday are made on Tuesday or Wednesday — and those decisions require demand intelligence that most planning teams don't have in a timely, actionable form by midweek.
Why Allocation Timing Is a Data Availability Problem
The fundamental challenge with DC-to-store allocation timing isn't logistics — it's decision intelligence. Deciding which SKUs to move from the DC to which stores before a weekend requires answering three questions:
- Which stores are tracking toward a stockout on which SKUs before Sunday?
- Of those at-risk SKUs, which have available units at the DC that can ship and arrive before Friday?
- Which reallocation will produce the highest demand capture — i.e., among multiple at-risk stores, which should get priority for the available units?
Answering question one requires current store-level POS data and a near-term demand forecast by SKU and location. Most omnichannel brands have the POS data; many don't have the localized demand forecast. They know what's selling now, but they don't have a tool that projects which stores are heading toward a weekend stockout based on current velocity and current on-hand.
Answering question two requires real-time DC inventory visibility. This is generally available but often siloed in a system that isn't connected to the planning workflow.
Answering question three requires a prioritization logic — and this is where most manual processes break down. When the planning team has to answer these three questions manually, for a catalog of 100+ SKUs across 6+ store locations, on a Tuesday afternoon with limited time, the result is usually a best-guess allocation that misses the highest-impact opportunities.
Building the 48-Hour Allocation Cadence
The operational target is a Tuesday-to-Thursday allocation cycle: demand signals pulled and analyzed Tuesday, allocation decisions made and transfers initiated Wednesday, merchandise in transit Thursday morning for Friday morning store receipt.
For brands not yet running this cycle systematically, the first step is establishing the data pull cadence. Pull store-level on-hand and trailing-2-week velocity for all SKUs on Tuesday morning. Calculate projected weeks-of-stock at each store as of next Monday (5 selling days out). Flag any store-SKU combination where projected WOS is below your threshold — typically 1.0–1.5 weeks triggers an allocation review for a weekend-peaking SKU.
The second step is cross-referencing against DC availability. Which of the flagged store-SKU pairs have available units at the DC (above your DC buffer minimum)? That's the reallocatable pool.
The third step is prioritization: among stores competing for limited available units, allocate to the highest-velocity stores first. The store that will exhaust a SKU by Saturday afternoon should get priority over the store that will exhaust it by next Tuesday, all else equal.
A Scenario: Six Stores, One SKU, One DC
A home goods brand running six retail locations and a DTC operation, managing approximately 140 SKUs. Their bestselling candle trio SKU runs at different velocities by location: their urban flagship sells 22 units/week, two suburban locations sell 12–14 units/week, and three smaller outpost stores sell 4–6 units/week.
On a Tuesday pull, the data shows: flagship has 18 units on hand (WOS = 0.8 weeks), suburban location 1 has 9 units (WOS = 0.7), suburban location 2 has 31 units (WOS = 2.4), outpost stores have 6–12 units on hand (WOS = 1.2–2.4). DC has 84 units available above the buffer minimum.
The allocation decision is clear: the flagship and suburban location 1 need units before Friday. The flagship gets priority — higher velocity means a stockout there costs more revenue. Allocating 20 units to the flagship and 15 units to suburban location 1 brings both locations to approximately 1.7 WOS after the weekend, adequate coverage. The remaining 49 DC units stay available for DTC and for the next allocation cycle.
Without a structured Tuesday review, this brand would likely have run through the weekend at both high-velocity locations, left the low-velocity suburban location 2 with its excess 31 units untouched, and shipped on Monday to cover the shortfall — missing two to three days of sales at the locations that could have moved the most volume.
The Antipattern: Reacting to Stockouts Instead of Preventing Them
Many brands run their DC-to-store allocation on a reactive basis: a store calls or emails to say they're out of a SKU, the DC ships units, the units arrive 2–3 days later. This is fine for preventing persistent stockouts, but it doesn't capture the weekend demand spike — by the time the reactive replenishment arrives, Saturday and Sunday have already happened.
We're not saying reactive allocation is useless — it has its place in keeping stores stocked over longer horizons. But reactive allocation alone leaves the highest-demand selling window entirely to chance. The combination of a weekly proactive allocation cycle (Thursday prep) plus a reactive replenishment system (covering missed stockouts) gives brands both the peak-capturing and the steady-state coverage they need.
When the 48-Hour Window Compresses
The 48-hour cycle assumes a transit time of roughly 24 hours from DC to store — reasonable for a regional distribution model where stores are within one-day shipping distance of the DC. For brands with geographically distributed networks where some stores are two or three transit days from the DC, the planning window needs to extend accordingly.
For a Monday receive, you're allocating Thursday. For a Tuesday receive, you're allocating Friday of the prior week. The logic is the same; the lead time inputs change. The key is knowing your actual DC-to-store transit times by location and building the allocation cadence around that reality rather than assuming all stores have the same replenishment speed.
The brands that execute this well have translated a seemingly logistical question — when does inventory move? — into a strategic one: which sales do we capture because we moved the right inventory at the right time? That reframe changes how planning teams think about allocation. It's not a logistics coordination task. It's a revenue capture decision.
