ABC Analysis for Inventory: Put Your Control Where the Value Is

1What ABC Analysis Actually Classifies

ABC analysis is a sorting job with a purpose. You take every SKU, score it on a single number — the value it moves through your business in a year — rank the whole list high to low, then slice it into three bands. The A band is the handful of lines at the top that carry most of your value. The C band is the long tail at the bottom that carries very little. B sits in the middle. That’s the entire structure.

The point isn’t the labels — it’s what they let you stop doing. Once a line is a C, you’re allowed to ignore it most of the time. Once a line is an A, it earns the attention you were previously smearing across your whole catalogue. The classification exists so that finite effort — counting, reordering, chasing suppliers — lands where a mistake actually hurts.

Crucially, this is not the same job as counting stock or deciding when to reorder. ABC tells you which lines matter most; the counting method and the reorder trigger are how you then treat each class. Get the classification right first, and every other stock decision inherits a sensible priority.

2Usage Value, Not Price Tag

The number that decides a line’s class is annual usage value: unit cost multiplied by the number of units you move in a year. This trips people up, because instinct reaches for the price tag. A £900 machined part feels like an A. But if you sell four a year, its annual usage value is £3,600 — while a 60p fastener you shift 40,000 times a year moves £24,000. On value, the fastener is the A and the expensive part is nearer a C.

That’s the insight most gut-feel stock control misses. Value flows through cheap, fast-moving lines just as surely as through expensive slow ones — often more so. Rank on usage value and the true shape of where your cash lives shows up, sometimes upside-down from where everyone assumed it was.

One inventory manager we spoke to described running the numbers for the first time and finding their “premium” SKUs — the ones the team fussed over — sitting quietly in the C band, while a boring consumable nobody thought about turned out to be the single biggest chunk of annual value in the building. Nothing had changed on the shelf. What changed was where they knew to look.

3Running the Classification

The mechanics are dull and that’s a good thing. Pull a line-level list: every SKU, its unit cost, and its annual units sold or consumed. Multiply cost by volume for each to get annual usage value. Sort the whole list descending. Then run a cumulative percentage down the column — what share of total value you’ve accounted for by the time you reach each line.

Now draw the bands off that cumulative curve, not off the raw list. Common cut-offs: the top lines that make up roughly the first 80% of value are your A class, the next lines up to about 95% are B, and the remaining tail to 100% is C. In a typical catalogue that lands somewhere near 20% of SKUs in A, 30% in B, and 50% in C — but the percentages are yours to set, not gospel. A distributor with a flat catalogue and a manufacturer with a few dominant components will draw the lines in different places.

Two practical notes. First, use a full year of data or you’ll misread seasonal lines. Second, decide upfront how you handle brand-new SKUs with no history and one-off spikes — they distort a raw sort. Park them, flag them, or annualise a partial period, but don’t let a single freak month promote a nothing line into A.

4What You Do Differently for Class A

Class A is where the money is, so it gets the tight control — and “tight” means specific things. Count A lines often, because an error on an A line is an error on real cash; a monthly or even weekly inventory cycle count on A items catches drift long before an annual stocktake would. Keep buffers lean and deliberate — you don’t want a mountain of your most expensive value sitting idle, so A lines want a sharp, well-tuned reorder point system rather than a fat just-in-case cushion.

Watch supply on A lines like a hawk. These are the lines where a stockout costs you a sale you can’t afford to lose and a supplier slip hurts most, so lead times, reorder points and supplier reliability all get close, named attention. If any line justifies a human actually thinking about it, it’s an A.

The trade is deliberate: you spend disproportionate effort on a small number of lines because that’s where a small percentage improvement returns real money. Tighten counting accuracy and reorder timing on the 20% of lines carrying 80% of value, and you’ve moved the number that matters.

5What You Do Differently for Class C

Class C is the opposite discipline: do less, on purpose. These lines carry so little value that the effort of managing them tightly costs more than the stock itself. So you count them rarely — once or twice a year is often plenty — carry generous buffers so you’re not reordering constantly, and order in bigger, less frequent batches. The goal is to spend as little management attention as possible per C line.

The mistake here is applying A-grade rigour to C-grade stock. A team that cycle-counts every washer weekly, chases a 20p line’s stockout, and reorders consumables in tiny careful batches is burning expensive hours protecting cheap value. Loosen the grip on C deliberately. A bigger buffer on a trivial line isn’t waste — it’s cheaper than the labour of managing it tight.

One caveat that keeps C sane: loose control is not no control. C lines still drift into dead stock if nobody ever looks — a C that stops selling entirely is a different problem from a C that sells slowly. Loose means low-frequency, not never.

6Class B and the Danger of Over-Engineering

Class B is the middle, and the honest answer is it gets middle treatment — moderate count frequency, sensible buffers, standard reordering. The temptation is to build elaborate rules for B, but the more useful move is to watch which way B lines are drifting. A B climbing toward A behaviour should get promoted before it stockouts; a B sliding toward C should get demoted before you overstock it.

Here’s the OpsMavix contrarian take: three classes is a tool, not a religion. Plenty of businesses over-engineer this into five or seven tiers with fractional cut-offs and lose the entire point, which was to simplify where attention goes. If your team can’t state a line’s class and what that means for how they treat it in one sentence, the scheme is too clever. Three buckets people actually act on beats seven buckets nobody remembers.

The value of B is mostly as a holding zone — a place lines pass through on the way up or down. Treat the boundaries as live, not fixed, and B does its job.

7Classes Drift — Why a One-Off Sort Rots

The quiet failure of ABC analysis is doing it once. You run the sort, colour the spreadsheet, feel organised — and six months later a line that was a solid A has faded to a C, a new product has climbed into A unnoticed, and your control effort is still pointed where the value used to be. Demand moves, costs move, products launch and die. A static classification describes a business that no longer exists.

The fix isn’t to re-run the sort by hand every quarter, because in practice nobody does — it’s a job that always loses to something more urgent. The fix is to make the classification live: re-rank continuously as sales and costs update, so a line’s class reflects this month’s reality, and flag the promotions and demotions so a human sees them. A line crossing from B into A is a signal worth acting on the week it happens, not the next time someone remembers to rebuild the sheet.

That’s the difference between ABC as a one-off tidy-up and ABC as a working control layer. The classification only earns its keep if it’s current — an inventory automation system keeps every line’s class trued up to live data so the effort always follows where the value actually is now.

FAQ

Q: What is ABC analysis in simple terms?

It’s sorting your stock into three groups by how much value each line moves in a year, so you can put tight control on the small number of lines that carry most of your cash (A) and loose control on the large number that carry very little ©. It’s the Pareto 80/20 principle applied to inventory.

Q: How do I calculate which class a SKU belongs to?

For each SKU, multiply the unit cost by the annual units sold or used to get its annual usage value. Rank every line high to low on that number, run a running total of the percentage of value, then cut the bands: roughly the top 80% of value is A, the next 15% is B, the final 5% is C. The class comes from usage value, not the price tag.

Q: Is ABC analysis the same as a reorder point or a cycle count?

No — and mixing them up is common. ABC classifies which lines matter most. The reorder point decides when to buy a line, and cycle counting is how you keep counts accurate. ABC sets the priority; the other two are how you then treat each class. A lines get frequent counts and tight reorder points, C lines get rare counts and generous buffers.

Q: How often should I re-run ABC analysis?

As often as your data updates, ideally continuously. Classes drift as demand and costs move, so a sort you run once and forget will point your effort at last year’s value within a couple of quarters. If you’re doing it by hand, quarterly is a realistic minimum — but a live system that re-ranks automatically and flags promotions and demotions is the version that actually stays useful.

Q: What are the right percentage cut-offs for A, B and C?

There’s no universal law — 80/15/5 of value (landing near 20/30/50 of SKUs) is a common starting point, but it depends on your catalogue. A business with a few dominant lines will have a steeper curve than one with a flat spread. Draw the bands off your own cumulative-value curve where the gaps naturally fall, and adjust as you learn what each class means for your workload.

How OpsMavix Can Help

Most of the businesses we work with don’t have an ABC problem on paper — they have it in practice. Effort is spread evenly because that’s how the spreadsheet grew, so cheap lines get babysat and the real value drifts unwatched. Running the classification once in a spreadsheet is easy; keeping it true as your sales and costs move, and actually changing how each class is counted and reordered off the back of it, is where it falls apart. That’s the gap we build across.

We build custom inventory systems that rank every line by live usage value, keep each SKU’s class current as the numbers change, and wire the classification straight into how you count, buffer and reorder — tight where the value is, loose where it isn’t — with no more manual re-sorting. If your team is guarding £2 boxes while your best lines get a glance, the first step is seeing exactly where the effort is landing wrong. Book a Free Operations Leak Audit.