INVInventoryStop 1 of 11
Why inventory exists
Cycle, safety, pipeline and anticipation stock, and what each one costs you to hold.
Read atChange any time
Inventory is stock you hold today so that tomorrow’s need can be met straight away. Companies hold it because supply and demand never move in step: suppliers ship in batches, factories run long batches, and customers buy in small, unpredictable amounts. Stock is the buffer that lets each stage run at its own pace.
Almost every unit on a shelf exists for one of four reasons:
- Cycle stock comes from ordering in batches. A café that buys 77.5 kg of coffee beans at a time slowly works through each batch, so on average it holds about half a batch.
- Safety stock covers surprises: a busier week than forecast, or a truck that arrives two days late.
- Pipeline stock is on its way: on a ship, a truck or a production line. If 1,000 units a day spend 6 days at sea, about 6,000 units are always in transit, even though nobody can sell them yet.
- Anticipation stock is built ahead of a peak you can see coming, such as sweets before Diwali or fans before summer, because the factory cannot make enough during the peak itself.
None of it is free. Stock ties up cash that could be used elsewhere, needs space and insurance, and can be damaged, stolen or go out of date. Many companies put the yearly cost of holding an item at 20–30% of its value. Too little stock costs money too: lost sales, idle production lines and expensive rush deliveries.
Words to use
- Cycle stock
- Stock that exists because you order or produce in batches; on average about half the batch size.
- Pipeline stock
- Stock in transit or in process. Average pipeline = flow per day × days in the pipeline.
- Holding (carrying) cost
- The yearly cost of keeping one unit in stock: capital, storage, insurance, damage and obsolescence.
Sizing each kind of stock
Each of the four reasons has its own simple formula, which is why naming the reason is the first step to managing it.
| Kind | Rough size | What drives it |
|---|---|---|
| Cycle stock | , half the order quantity | Batch size |
| Safety stock | Uncertainty and the service target | |
| Pipeline stock | , demand rate times transit time | Lead time |
| Anticipation stock | Demand above capacity, built in advance | Seasonality and capacity limits |
Pipeline stock is an example of Little’s law, one of the most useful relationships in operations: average inventory = throughput × average time in the system. It holds for a warehouse, a factory or a ship. It also means the only ways to cut pipeline stock are to sell less (rarely the plan) or to shorten the time goods spend in the pipe.
A fan maker can build 25,000 fans a month. Demand is 10,000 a month from January to March, then 40,000 a month from April to June.
- April to June needs 120,000 fans, but the plant can build only 75,000 in those months.
- The 45,000 gap must be built in advance. Running at full capacity from January to March adds 25,000 − 10,000 = 15,000 fans a month to stock: exactly 45,000 by the end of March.
- That stock is sold down through the summer. It exists only because capacity is flat and demand is not.
Measuring the total: turns and days
Inventory is usually tracked with two linked numbers:
A distributor with cost of goods sold of ₹120 crore a year and average inventory of ₹20 crore turns its stock 6 times a year, about 61 days of inventory. Value both at cost, not at selling price, or the ratio is inflated.
What holding stock costs
The yearly holding cost is usually expressed as a percentage of inventory value. It is built from several parts: the cost of the capital tied up, storage and handling, insurance and taxes, and losses from damage, theft and obsolescence. At 25%, the distributor above spends about ₹5 crore a year just to keep its ₹20 crore of stock. That number is what every inventory decision is weighed against.
Answering “cut inventory by 20%”
An across-the-board cut is the worst answer, because it treats all stock as if it existed for the same reason. Break the total into its four kinds plus excess, then attack each driver:
| Kind of stock | Lever that reduces it | Watch out for |
|---|---|---|
| Cycle stock | Smaller, more frequent orders; cheaper set-ups and ordering | More orders, trucks and changeovers |
| Safety stock | Better forecasts, more reliable suppliers, shorter lead times, service targets by segment | Cutting buffers on volatile items destroys service |
| Pipeline stock | Faster modes, nearer suppliers, shorter internal flow times | Faster freight usually costs more |
| Anticipation stock | More flexible capacity, smoother promotions | Overtime and extra shifts cost money |
| Excess and obsolete | Stop buying, sell off, write down | A one-off profit hit |
The quickest win is usually excess and obsolete stock, which has no reason to exist at all. The most durable wins come from shrinking the drivers themselves: lead times and forecast error.
Moving the curve, not sliding along it
For a given set of lead times, forecast errors and batch sizes, there is a trade-off curve between inventory and service. You can slide along it, accepting worse service for less stock or the reverse. That is a choice, not an improvement. Real improvement shifts the whole curve, so the same service needs less stock; halving lead-time variability or forecast error does exactly that. When someone proposes an inventory cut, ask which of the two they mean.
Stock hides problems
Lean thinking compares inventory to the water level in a river: high water hides the rocks. A plant with plenty of buffer never notices that one machine breaks down every week or that one supplier is always late. Lowering stock deliberately, a little at a time, exposes those problems so they can be fixed. Lowering it suddenly, without fixing anything, runs the boat onto the rocks.
Where stock sits matters as much as how much
The same rupee of inventory is more or less useful depending on where it sits. Raw materials are cheaper to hold and can become many products, but they are slow to turn into a sale. Finished goods near customers sell instantly, but only as one product in one place. The point where a product stops being made to forecast and starts being made for a specific order is the customer order decoupling point. Make-to-stock, assemble-to-order and make-to-order businesses differ mainly in where that point sits.
Interview kit
Practise it out loud
Answer each question yourself before opening the model answer.
Companies hold inventory because supply and demand don’t match in timing or size. Cycle stock comes from ordering in batches, safety stock covers uncertainty, pipeline stock is what’s in transit, and anticipation stock is built ahead of a known peak. Each type has a real holding cost, often 20–30% of the item’s value a year, so the aim is to hold each type only for the reason it exists.
A CFO asks you to cut inventory by 20%. Where do you start?
Split the inventory by type before cutting anything. Cycle stock falls with smaller, more frequent orders, if ordering costs allow it. Safety stock falls with better forecasts and more reliable suppliers. Pipeline stock falls only if transit or process time gets shorter. Obsolete stock should be cleared. A blanket cut usually just removes safety stock and causes stockouts.
What is pipeline inventory, and how would you estimate it?
It is stock in transit or in process. On average it equals the flow rate times the time spent in the pipeline: 1,000 units a day with a 6-day transit means about 6,000 units in transit. You reduce it by shortening the lead time, not by ordering less often.
Is inventory good or bad?
Neither on its own. On the balance sheet it is an asset, but operationally it ties up cash and can hide problems such as unreliable suppliers or long changeovers. Inventory held for a clear reason is useful; inventory nobody can explain is waste.
Connected stops
Where this idea feeds in, or gets its inputs from.
Sources and further reading
- Chopra, S. & Meindl, P. Supply Chain Management: Strategy, Planning, and Operation. Pearson. Chapters on cycle inventory and safety inventory.
- Silver, E. A., Pyke, D. F. & Thomas, D. J. (2017). Inventory and Production Management in Supply Chains (4th ed.). CRC Press.
- Little, J. D. C. (1961). A proof for the queuing formula L = λW. Operations Research, 9(3), 383–387.
- Hopp, W. J., & Spearman, M. L. (2008). Factory Physics (3rd ed.). McGraw-Hill.