For every one of hundreds of SKUs: when to reorder and how much — the trade-off between running out and tying up cash in stock (known in the literature as the (s,S) policy).
In plain words
Sound familiar?
- 'When did this product run out?' has no clear answer — you find out at the next stock count
- Some products sit in the warehouse for 6–12 months; cash is locked there while suppliers chase invoices
- You hit 'customer asked, we were out' 3–8 times a month, and a fraction of those customers don't come back
- Rush orders cost extra freight or a supplier markup — and they break the budget
- The supplier MOQ is high; for slow movers you over-order just to satisfy the minimum
- Going into a season, 'how much of each SKU should we hold' relies on gut feel and broken last-year data
- Expiring SKUs get discounted or written off — your shrinkage rate sits at 3–8%
Why it matters
How it's solved
Technical depth
How it's solved
Technical depthOne-liner: Pick two numbers per product — a reorder level (place an order when stock falls to this) and a fill level (the order should bring stock back up to here). The more volatile the demand and the slower the supplier, the higher the reorder level needs to be — otherwise you stock out.
What the software is really doing is this: the same when/how-much decision your buyer makes from memory for 50 SKUs, it makes for 5,000 in seconds, and updates every night. Three stages:
1. It collects the data. For each SKU: 6–12 months of sales history, supplier lead time and lead-time variability, MOQ and price breaks, shelf life if any, warehouse space limits, and a target service level (e.g. 95% — ninety-five out of a hundred customers find what they came for). The data flows from POS, e-commerce, or ERP automatically, or is entered once into a clean table.
2. It computes the policy for each SKU. The software does not look at every SKU every day — there is no mathematical need. Instead, it uses inventory modeling — a body of work from operations research (a discipline that uses math and computing to solve business-decision problems) — to compute two numbers per SKU: s (the reorder point — when stock falls to this, place an order) and S (the order-up-to level — restock to this). These two numbers are derived from demand variability, lead time, and the service level you want. When demand, lead time, or supplier terms shift, the software updates the numbers automatically.
3. The reorder list lands on the desk each morning. The software shows which SKUs to order today, from which supplier, in what quantity. You say ‘yes, send,’ and the system forwards the order to the supplier and starts tracking the delivery. If the supplier slips or changes MOQ, the software flags it.
It does not replace your judgement; think of it as a calculator that scales what you do for 50 SKUs up to 5,000 and never forgets a single one. The decision is still yours, but you always have a current table for every product."
Alternatives
Spreadsheet + buyer's intuition
FreeFree
Who it fits: 10–100 SKUs, stable demand, long shelf life
- + Zero cost
- + Flexible — easy to override on intuition
- + No capex decision
- − Quality drops fast above ~100 SKUs
- − Demand variability and lead-time uncertainty aren't accounted for
- − Season turns surprise the buyer
- − Cash tied up per SKU isn't visible
ERP inventory module
Enterprise$500–2,000 setup + $100–400/month (regional SMB pricing)
Who it fits: 100–1,000 SKUs, stable supplier base
- + Local-language interface and support
- + Wired into accounting and invoicing
- + Automated reorder-point alerts usually present
- − Demand forecasting is usually 'last month × 1.1' simple
- − Service level / safety stock math is missing or rudimentary
- − Multi-echelon (warehouse + stores) optimization is weak
International specialized inventory and supply software
Enterprise$5–25/SKU/month subscription, or $30,000–200,000/year licence
Who it fits: 1,000–50,000 SKUs, multi-warehouse, demand-forecast-driven planning
- + Mature: statistical demand forecasting, multi-echelon optimization, alerting all fully supported
- + Supplier integration (EDI, API) included
- + Separates seasonality from underlying trend
- − High licence and consulting cost
- − Rollout takes 3–6 months
- − Local-language support can be limited
Custom build on an open-source solver
Open SourceLicence free; 8–16 weeks of internal development, or $50,000–250,000 of consulting
Who it fits: Chain with a data-science team, or an e-commerce business with unusual constraints
- + No licence cost
- + Fully customizable to your category structure
- + Cloud or on your own server
- − Requires real data and OR capacity in-house
- − Ongoing maintenance — every supplier change matters
- − Forecasting and policy are two separate disciplines; building both teams is expensive
Recommendation
Ask in the meeting
- What model underpins your demand forecast — simple average, seasonal decomposition, machine learning? At which scale does each kick in?
- How does lead-time variability feed safety stock? Can the service-level target be set per SKU?
- Are minimum order quantities (MOQ), lot sizes, and price breaks supported?
- How does the system model multi-echelon (warehouse + multiple stores, or warehouse + e-commerce)?
- How is forecasting handled for new products with no history? Is there similar-product matching?
- For SKUs nearing expiry, does the system propose automated action (discount, transfer, halt ordering)?
- How do you structure the pilot — how many SKUs, how many weeks, what is the success bar?
- If we stop working with you, how do we get our product, supplier, and movement data back? Is there a standard export format?
Technical details
Editor’s note
On the buyer’s desk this problem is known as “stock replenishment”, “order management”, or “what goes on the shelf next”. The academic name is Inventory Replenishment with an (s,S) policy — the reorder-point + order-up-to-level rule. It has been one of operations research’s oldest pillars since the 1950s. Without that vocabulary you cannot tell whether the “inventory module” pitched to you actually does statistical demand forecasting and safety-stock math, or whether it’s a glorified low-stock alert.
The point most often overlooked in this segment: many products advertise ‘inventory management’ but underneath they only run physical count + alert threshold logic — i.e. “alert me when stock drops below 10.” That works when demand is flat; once demand swings or lead times vary, plan quality collapses. In any demo, ask the vendor to walk through a 20-SKU example where each SKU has different demand variability and different supplier lead times, and explain how the reorder point is computed per SKU.
A step-by-step path for an SMB
Stage 1 — Measure first, plan later. For at least 12 weeks, log four things in a spreadsheet:
- Daily/weekly unit sales per SKU (from POS or manual count)
- Order-to-delivery time per supplier (mean and variability)
- Number of stockouts and the cause (forecast miss, supplier slip, sudden campaign)
- A list of SKUs sitting 90+ days in the warehouse, and the cash tied up
Without this baseline you cannot tell which software will deliver which result.
Stage 2 — Do an ABC analysis. Roughly 80% of revenue comes from ~20% of SKUs (the A group). A-group SKUs need a tight policy; B-group medium; C-group simple. Do this categorization first yourself, then explain it to the software — automatic ABC is fine, but adding your operational view makes the result better.
Stage 3 — Pilot. Start with 30–50 A-group SKUs for 8–12 weeks. Define the success criterion in writing, before the pilot: e.g. “in 90 days, A-group stockouts down 50%, average inventory down 15%”. If the bar is missed, the pilot ends — keep that exit right in the contract.
Stage 4 — Rollout. If the pilot lands, scale to the whole SKU range over 2–4 months. Operations training runs 1–2 weeks; supplier integration (EDI or API) typically takes 4–8 weeks.
Risks — what can go wrong
- Bad data. If historical sales include stockout periods, demand forecasts come out too low — the software orders too little — a vicious cycle. The history needs to be ‘censored-stockout-corrected’; a good system does this itself.
- Operations team resistance. “The computer says order 50 units but my gut says 20” is common. In the pilot, walk through the result with the team; the software must show transparently which rule was applied.
- Supplier pushback. Smaller, more frequent orders can hit supplier MOQ or freight pricing. Renegotiating the supplier contract should be part of the pilot scope.
- Vendor lock-in. Software that stores movement history and supplier data in a proprietary format makes migration hard. Put a clause in the contract: “We can export our data in standard open formats (CSV or similar) at any time, on request.”
Related cautionary lesson (will be linked once published): “A 5,000-SKU retailer that dropped its inventory software at month 14 — what they missed.”
A technical view of the solution method
This section holds what you’ll need when talking to a software team or a consultant. It is not what the operations team sees in their daily screen — it is the engine behind the curtain.
The main approaches used for inventory replenishment:
| Approach | Typical use | Data need | Decision logic |
|---|---|---|---|
| EOQ formula | Stable demand, single product | Low | One formula, instant compute |
| (s,S) policy | Variable demand, many products | Medium | Stochastic model + safety stock |
| (R,Q) policy | Fixed-cycle review days | Medium | Periodic review |
| Newsvendor / single-period | Perishable, seasonal | Medium | Demand distribution + cost ratios |
| Multi-echelon MIP | Warehouse + store network | High | Full network optimization |
In practice: under 500 SKUs with stable supplier terms, (s,S) or (R,Q) is enough. Above 1,000 SKUs or in multi-echelon networks, commercial demand planning + multi-echelon optimization is preferred. For perishable goods, newsvendor variants are used.
Objective function choice changes the shape of the solution:
- Total carrying cost: “Minimize trapped cash” — fits a cash-tight business
- Service level (e.g. 95% fill rate): “No customer leaves empty-handed” — fits a market-share-driven business
- Expected profit (including lost sales and write-offs): “Maximize total profit” — fits a high-margin retailer
- Stockout cost + write-off cost balance: “Perishable strategy” — fits fresh food or pharma
Most real deployments use a weighted blend of all four.
Academic references
Listed in the sources block of this page’s frontmatter. Inventory management has been one of operations research’s oldest active fields (since the 1950s); INFORMS Interfaces and the European Journal of Operational Research archive carry deployment case studies tied to real retail and supply chain operations.
Sources
- Silver, E. A., Pyke, D. F. and Thomas, D. J. (2017). Inventory and Production Management in Supply Chains (4th ed.). CRC Press. The standard textbook in the field.
- Zipkin, P. H. (2000). Foundations of Inventory Management. McGraw-Hill. The mathematical reference for single- and multi-period inventory models.
- Nahmias, S. and Olsen, T. L. (2015). Production and Operations Analysis (7th ed.). Waveland Press. A presentation close to SMB practice.
- INFORMS Interfaces — case studies of operations research deployments in retail and supply chain. informs.org/Publications/Interfaces
Glossary
- EOQ
- The classic inventory formula for the most economic order quantity to place with a supplier.
- Safety Stock
- Extra stock held against demand and lead-time uncertainty — protects against stockouts during unexpected swings.
- MIP
- An optimization model where some decision variables are forced to be whole numbers (e.g. number of trucks, number of shifts).
Related problems
Dozens of SKUs from the Same Supplier — At What Frequency Do I Order Each So Trucks and Stock Cost Are Minimum Together?
This page is for an SMB wholesaler, importer or manufacturer pulling 50-500 SKUs from the same supplier — typically across 3-15 main suppliers. The weekly question is the same: from this supplier, which products should ship today and which can wait until next week? If every product triggers its own order, the same supplier ends up sending three trucks and three customs filings a week; one truck, one customs filing and one setup cost can cover all of it if the groups are right. Manual planning falls apart past 50 SKUs: some weeks a half-empty truck, others three back-to-back orders — fixed costs come back as 5-15% of unit product cost.
How Much of Each Product Should I Order So I Neither Run Out Nor Pile Up?
Written for retail, e-commerce, or manufacturing SMBs managing 500-5,000 SKUs. If your purchasing team still places orders every month from a spreadsheet using last-three-months average plus ten percent, and yet you see empty shelves and clogged warehouses at the same time, you are in this problem. When the same company carries both stockouts (5-15% lost sales) and dead inventory (10-25% capital tied up), intuition has hit its limit: seasonality, lead-time uncertainty, and slow-moving SKUs cannot be corrected by hand item-by-item. Systematic forecasting paired with an inventory policy typically frees 1.8-5.4M TRY/year of working capital on 50M TRY revenue and lifts the service level from roughly 85% to 95%.
Stock Is Not Clearing as the Season Ends — When and How Deep Should I Mark Down So That Margin Holds and Dead Stock Stays Low?
This page is for you if you run a seasonal-collection fast-fashion retailer, a 30-100 store apparel/shoes/accessories chain, or a fresh-food chain (greengrocer, bakery, butcher). The classic pain: a 200-1,000 SKU collection sells for 4-12 weeks and then the unsold stock becomes a problem — discount too early and you erode the gross margin, discount too late and you sit on a pile of dead stock at the end of the season. The decision: which item, when, and by what percentage to mark down? In a mid-size fashion chain run on gut-feel rules, 20-35% of end-of-season stock ends up as salvage; a proper markdown schedule lifts the gross margin by 5-15%, cuts dead stock by 20-30%, and adds 15-60 million TRY of operating margin a year on a typical revenue base.