Replenishment

ABC analysis for inventory, and where the letters stop helping

Attention is the scarce resource, not stock. ABC analysis is the standard way to ration it, and it is right about the problem and wrong about the day.

Marcus VolstedCo-founder, Restocio Published 6 September 2026Updated 6 September 20267 min read

ABC analysis for inventory, and where the letters stop helping

ABC analysis ranks your products by how much money passes through them in a year, then splits them into three classes so you can spend your attention on the ones that matter. The premise is correct and most stores need it. The limitation is that annual value is a description of last year, and reordering is a question about this week.

Both things can be true. It is a good tool used at the wrong moment more often than it is a bad tool.

What is ABC analysis?

Pareto applied to a catalogue. Roughly 20 percent of your products carry roughly 80 percent of the value, so you rank everything by annual consumption value and cut the list into three groups.

A items carry the first 80 percent of total value, and usually turn out to be about 20 percent of your products. Tight control, frequent review.
B items carry the next 15 percent. Normal handling.
C items carry the last 5 percent, and are often half your catalogue by count. Loose control, bigger and rarer orders.

How do you actually calculate it?

Multiply each product's unit cost by the units sold in a year, sort descending, and run a cumulative percentage down the list. Where the cumulative crosses 80 and 95 percent is where your classes break.

Product	Unit cost	Annual units	Annual value	Share	Cumulative	Class
Product 1	42 kr	14 000	588 000 kr	52.5%	52.5%	A
Product 2	85 kr	4 200	357 000 kr	31.9%	84.5%	A
Product 3	19 kr	6 000	114 000 kr	10.2%	94.6%	B
Product 4	120 kr	400	48 000 kr	4.3%	98.9%	C
Product 5	8 kr	1 500	12 000 kr	1.1%	100%	C

It is a spreadsheet job and it takes an afternoon. Use cost rather than selling price if the question is how much capital a product ties up, which for a purchasing decision it usually is.

What does ABC get right?

The thing almost every store gets wrong, which is treating a catalogue as if every line deserved equal thought. Nobody reviews nine hundred products properly. What actually happens is that the products people remember get checked and the rest are sampled, and sampling without a rule means the rule becomes recency.

ABC replaces that with a defensible rule. That alone makes it worth doing once, and for two specific jobs it stays the right answer permanently: deciding how often to physically count a product, and deciding which suppliers are worth negotiating hard with. Both of those are genuinely annual questions, so an annual measure fits them.

Where do the letters stop helping?

Three places, and they compound.

The class is static and the decision is not. Most stores recalculate quarterly at best. A letter assigned in January says nothing about whether a product needs ordering in March, and the ordering decision is the one you make every week.

Value is not urgency. An A item with two hundred days of stock on the shelf needs nothing from you today. A C item that takes ninety days to arrive and has thirty days of cover left is about to cost you two months of sales. ABC ranks the first above the second, every single time, because the axis it sorts on has no idea when anything runs out.

In the table above, Product 4 is a C item on 400 units a year. That is a little over one a day, so thirty days of cover is barely thirty units on the shelf, and if it ships from Asia it will be unavailable for roughly two months. Product 1 is an A item and might be comfortably stocked until the summer. The letters point your attention at exactly the wrong one.

It is blind to the products that have already failed. This is the sharpest one. A product that has been out of stock for two months sold almost nothing during the window, so its annual consumption value collapses and it classifies as C. The ranking demotes a product because it failed, which is precisely backwards.

What should you rank by instead?

For the reorder decision specifically, by the money riding on what is actually due. Not what a product was worth last year, but what the units you need to order right now would sell for, which combines size and urgency in one number and recomputes itself every day.

A product that needs nothing this week scores zero regardless of how big it is. A product that is about to go dark scores its full replacement value. That ordering changes daily on its own, with no quarterly reclassification exercise, because it is derived rather than assigned.

How Restocio ranks products

By revenue at risk: the quantity the engine says is due, multiplied by what those units sell for excluding VAT. Products that need nothing fall to the bottom without anyone filing them there.

That measure has one failure and we handle it by exception, which is worth describing because it is the same trap as ABC's third blind spot. A product that has already run out has nothing due that can be valued against sales it is not making, so its risk value tends toward zero and a pure value sort would bury it. Products that are stocked out with no sales are therefore lifted out of the ranking and shown first, with an explicit out-of-stock state rather than a reassuring one. Any ranking built on value needs that exception, including yours.

What we do not do

We do not produce A, B and C letters, and there is no classification screen. If you need literal classes, because your counting schedule or your warehouse process is built around them, export your costs and sales and do it in a spreadsheet. It is an afternoon and then a quarterly repeat, and pretending a continuous ranking is the same thing would help nobody.

The two tools answer different questions. ABC tells you which products deserve a policy. A live ranking tells you which products need a decision today. A store past a few hundred products probably wants both, and only one of them has to be recalculated by hand.

Common questions

What is ABC analysis in inventory management?

A method that ranks products by annual consumption value, unit cost multiplied by units sold in a year, and splits them into three classes. A items carry the first 80 percent of value and are usually about 20 percent of the catalogue, B items the next 15 percent, and C items the last 5 percent, which is often half the products by count. The purpose is to ration attention rather than spread it evenly.

How do you calculate ABC analysis?

Multiply unit cost by annual units sold for every product, sort the list descending by that value, then run a cumulative percentage down it. Products up to 80 percent cumulative are A, up to 95 percent are B, and the remainder are C. Use cost rather than selling price when the question is how much capital each product ties up.

Should I use ABC analysis to decide what to reorder?

Not on its own. Annual value describes last year and a reorder decision is about the next few weeks, so the classes cannot see urgency. An A item sitting on two hundred days of cover needs nothing, while a C item with a ninety-day lead time and thirty days of stock is about to lose you two months of sales. Use ABC for policy, such as counting frequency and supplier negotiation, and something time-based for the ordering itself.

What is the biggest weakness of ABC analysis?

It demotes products that have already failed. A product that spent two months out of stock sold almost nothing in the measurement window, so its annual value collapses and it drops into class C, which means it gets less attention precisely because it ran out. Any ranking based on value needs an explicit exception for stocked-out products or it hides its own worst cases.

How often should you redo an ABC analysis?

Quarterly is the common answer and it is reasonable for the jobs ABC is actually good at, since counting schedules and supplier terms do not change weekly. If you find yourself wanting to rerun it monthly to keep ordering decisions current, that is a sign you need a measure that recomputes itself rather than a faster manual reclassification.

What is the 80/20 rule in inventory?

The Pareto principle applied to a catalogue: roughly 20 percent of your products account for roughly 80 percent of the value moving through the business. It is the idea ABC analysis formalises, and it is a rough observation rather than a law, so check it against your own numbers before designing a process around it. The ratio in your store might be 30/70 or 10/90.

Marcus Volsted

Co-founder, Restocio

Marcus co-founded Restocio and works on it daily with a Swedish importer who plans their purchasing in it every working day. Restocio is built in Sweden by two founders, one Swedish and one Danish, and Marcus is the Danish one. Most of the examples on this blog come from that store's real ordering decisions rather than from a textbook. Why we are building it.