What is inventory management?

A working definition

Inventory management is the set of practices by which a business tracks the goods it holds and decides how much of each to hold. That covers ordering, receiving, storing, counting, valuing and eventually selling or writing off — and it covers the decisions as much as the record-keeping.

The record-keeping half gets most of the attention because it is visible: the counting, the scanning, the reconciling. But an accurate count that nobody acts on is just an expensive hobby. The decisions are where the money is.

The four decisions it exists to answer

Strip away the vocabulary and every inventory system, from a paper ledger to a warehouse management platform, exists to answer four questions:

1. What do I have? The current position, by product and by location, accurate enough to promise to a customer.
2. When should I order more? The reorder point — the level at which replenishment must start if you are to avoid running out.
3. How much should I order? The order quantity, balancing the cost of ordering against the cost of holding — the domain of EOQ.
4. What should I stop holding? Dead, obsolete and expiring stock, which quietly consumes capital and space until somebody decides to deal with it.

A system that answers the first question and none of the others is a stock count, not inventory management. That distinction is precisely where most spreadsheet-based setups sit.

The terms that actually matter

Types of inventory

Not all stock is the same kind of asset, and treating it as one is why some businesses cannot explain their own numbers:

• Raw materials — inputs waiting to be used. Driven by production schedules.
• Work in progress — partly finished goods. Consumes capital and is easy to lose track of entirely.
• Finished goods — ready to sell. What most retail and distribution businesses mean by "stock".
• MRO — maintenance, repair and operations supplies. Never sold, so frequently unmanaged, and frequently the cause of expensive downtime.
• Safety stock — deliberate buffer, not surplus. The distinction is a decision you recorded, not a property of the goods.
• In transit — bought and paid for, not yet arrived. Forgetting this is a classic cause of double-ordering.

The core methods

Reorder point with a fixed order quantity

The workhorse for small businesses. Each product has a trigger level and an order quantity; crossing the trigger fires an order. Simple, robust and easy to automate. Combine a calculated reorder point with an EOQ quantity and you have a complete policy.

Periodic review

Rather than a trigger level, you review on a schedule — every Monday, say — and top up to a target. Better where a supplier has a fixed delivery day. It requires slightly more buffer, because you carry risk over the review interval as well as the lead time.

ABC analysis

Sort products by the value they represent and treat the classes differently. Usually about 20% of lines carry 70–80% of value. The method is here and it is the highest-leverage hour you can spend on a catalogue you have never analysed.

FIFO, LIFO and weighted average

Costing conventions that decide which cost is released when you sell. FIFO — oldest first — matches how physical goods should move and is the sane default for anything perishable.

Cycle counting

Counting a slice of the catalogue continuously rather than shutting down once a year for a full count. More on that here.

What holding stock really costs

The invoice price is the visible cost. The annual carrying cost is the one that decides whether a policy is sane, and it typically runs 20–30% of unit value per year:

• Capital — cash in stock is cash not in the business. If you borrow, this is your actual interest rate.
• Storage — rent, shelving, handling, utilities.
• Risk — theft, damage, obsolescence, expiry.
• Insurance and tax on the value held.

At 25%, ₹10 lakh of stock costs about ₹2.5 lakh a year to hold. That figure is what makes "just order more, to be safe" an expensive instinct — and what makes the effort of setting reorder points properly pay for itself.

When a spreadsheet stops being enough

Spreadsheets are genuinely good at inventory up to a point, and it is worth being honest about where that point is. You have passed it when any of these is true:

• Two people need to update stock at the same time.
• You have asked "who changed this, and why?" and could not find out.
• Someone sold something that was already promised to another customer.
• You cannot answer what margin you made on a category last quarter without an evening's work.
• Returns, part-deliveries or credit notes have to be handled "carefully" by one person who understands the sheet.
• The stock figure is routinely wrong and everyone has learned to check the shelf instead.

The failure is not that spreadsheets cannot hold the data. It is that they have no concept of a transaction — a quantity change with a type, a reason, a timestamp and a person attached. Without that, history cannot be reconstructed and reports cannot be trusted. The full comparison is here.

How to start, in order

1. Get the catalogue right. One row per genuinely distinct sellable item, with a stable SKU. Do this badly and everything downstream is unusable.
2. Do one honest full count. Every later number is built on this one. Do not skip it and do not adjust it to match the books.
3. Record movements as they happen. Not at the end of the day, not from memory. This is a habit change, and it is the hardest part of any implementation.
4. Run ABC analysis. Find the 20% that carries the value.
5. Set reorder points on the A items properly. Default the rest. Perfection on the long tail is not worth the hours.
6. Start cycle counting. A items monthly, C items annually.
7. Review quarterly. Demand drifts, suppliers change, and a policy set two years ago describes a business that no longer exists.