What Is a SKU? Stock Keeping Unit Explained (+ Examples)

A SKU is the unique code your business gives to each distinct product you sell, so you can tell one item apart from another and track it through your system. It’s pronounced “skew” and stands for stock keeping unit. Every size, colour and variant gets its own code.

So a small blue t-shirt and a large blue t-shirt are two different SKUs, even though they’re the “same” shirt.

SKUs are the foundation the rest of inventory management is built on. Without a unique code per product, you can’t track stock accurately, run a report, or tell your system which item just sold. They sound trivial, but a business that skips them — or builds them badly — pays for it later in confusion and errors. This guide covers what a SKU is, how to structure one, how it differs from a UPC or a barcode, and how to create your own without making the common mistakes.

What a SKU is and what it stands for

A stock keeping unit (SKU) is an internal, alphanumeric code a business assigns to each unique product or variant to identify and track it. “Internal” is the key word: you create your SKUs, they’re yours, and they only need to make sense within your own business.

That’s what separates a SKU from the codes printed on retail packaging. A SKU is your private label for a product. It lets you look up an item instantly, see how many you have, know where it’s stored, and track how well it sells — all from one short string of characters.

The anatomy of a SKU

A good SKU isn’t random — it’s a small code with meaning built in. The idea is that a person can glance at it and read off the product’s key attributes without looking it up.

Take a SKU like TSH-BLU-M-SS. Broken down, it might mean:

• TSH — product type: t-shirt
• BLU — colour: blue
• M — size: medium
• SS — style: short sleeve

Each segment describes one attribute, usually ordered from the broadest (category) to the most specific (variant). The exact structure is up to you — what matters is that it’s consistent, readable, and scalable enough to cover every product you’ll ever add. A well-built SKU tells a picker or a manager what the item is at a glance.

Ordering the segments broad-to-specific is more than tidy — it’s practical. When you sort your product list alphabetically by SKU, related items group together automatically: all your t-shirts sit next to each other, then within them all the blue ones, then the mediums. That grouping makes reports readable and stock easier to find. Get the order the other way round — variant first, category last — and your list becomes a jumble of unrelated items that happen to share a size.

One more distinction worth clearing up: people sometimes say “SKU” to mean the product and “SKU number” to mean the code. In practice they’re used interchangeably — the SKU is both the identifier and, loosely, the thing it identifies. Either way, it’s your internal reference, and that’s what separates it from every standardised code below.

SKU vs UPC vs barcode vs GTIN vs ASIN vs model number

This is where people get tangled, so here’s each identifier and how it differs from a SKU:

The two big distinctions to hold onto: a SKU is internal and yours, whilst a UPC is external and standardised — the same product has one universal UPC but a different SKU in every business that stocks it. And a barcode is not a number at all — it’s the scannable stripes that carry a number, which could be your SKU or a UPC. (The types of barcodes guide covers the formats.)

What are SKUs used for?

SKUs earn their place across the whole business:

• Inventory tracking — every stock count, movement and report keys off the SKU.
• Faster picking and fulfilment — a picker finds the exact variant by its code, not by guessing.
• Sales analysis — you can see precisely which variants sell and which don’t, down to the size and colour.
• Reordering — reorder points and forecasts are set per SKU.
• Multichannel selling — one consistent SKU keeps stock in sync across your website, marketplaces and shops.

How to create your own SKUs

You can build a solid SKU system in an afternoon. The key is to design the logic once and apply it consistently.

1. Start with the broadest attribute first. Lead with product category or department, then narrow down (type, then colour, size, style). This keeps related products grouped when you sort them.
2. Use short, readable segments. Abbreviations a human can decode (BLU for blue, not a random number). Two to four characters per segment is plenty.
3. Keep it consistent. Every product follows the same segment order and format. Inconsistency is what breaks a SKU system.
4. Separate the segments. A dash or hyphen between segments makes SKUs easy to read at a glance.
5. Assign a unique code to every variant. Each size and colour combination is its own SKU.

Common mistakes to avoid:

• Don’t start a SKU with a zero. Spreadsheets strip a zero at the start of a number and quietly corrupt your data — SKU 0123 becomes 123, and now two products can collide.
• Don’t use letters that look like numbers. The letter O and the number 0, the letter I and the number 1, are a constant source of miskeyed and misread codes. Leave the ambiguous characters out.
• Don’t just reuse the manufacturer’s model number. It wasn’t built for your catalogue, it may change without warning, and two suppliers can use the same one. Your SKU is yours to control.
• Don’t overload it. A SKU that tries to encode ten attributes becomes unreadable. Capture the handful that matter (category, colour, size, style) and stop.
• Don’t recycle old SKUs. When a product is discontinued, retire its code — don’t hand it to a new product, or your historical sales data becomes meaningless.

Simple, consistent and unique beats clever every time. If you have to choose between a SKU that’s clever and one a tired picker can read at the end of a shift, choose readable.

Keeping SKUs consistent across sales channels

If you sell in more than one place — your own website, a marketplace, a physical shop — your SKU becomes the thread that ties your stock together. Use the same SKU for a product everywhere you sell it, and your systems can keep a single, shared count: sell one on Amazon and your website stock drops too.

Where businesses come unstuck is letting each channel invent its own code for the same item. Now the “same” product has three different identifiers, nothing reconciles, and you oversell because no single number reflects the true stock on hand. One product, one SKU, everywhere — that discipline is what makes multichannel selling manageable rather than chaotic. (Note that a marketplace may also assign its own identifier, like Amazon’s ASIN — that’s fine; it sits alongside your SKU rather than replacing it.)

Smart SKUs vs sequential SKUs

There are two schools of thought on how to build a SKU, and it’s worth knowing both before you commit.

A smart (or intelligent) SKU encodes meaning, like the TSH-BLU-M-SS example above. A person can read the code and know roughly what the product is. The upside is human-readability: pickers, buyers and managers can decode an item at a glance, and related products sort together. The downside is that smart SKUs get long as you add attributes, and if a product’s attributes change you can be tempted to change its code — which you shouldn’t.

A sequential (or non-smart) SKU is just a number assigned in order — 1001, 1002, 1003 — with the meaning held in your system, not the code. The upside is simplicity: codes stay short, never run out, and never need decoding logic. The downside is that the code itself tells a human nothing; you have to look every item up.

Most small and mid-sized businesses lean towards smart SKUs because the readability genuinely helps day to day. Very large catalogues, or businesses whose software does all the lookups anyway, often prefer sequential codes for their simplicity. Neither is wrong — the test is whether a human ever needs to read the code unaided. If yes, go smart; if the system always does the reading, sequential is fine.

When should you create a new SKU?

A common question is how far to take variants: does every tiny difference deserve its own SKU? The rule is simple — create a new SKU whenever a difference matters to you or your customer. If a customer can choose it, or you need to count it separately, it’s a new SKU.

In practice that means a new SKU for every distinct:

• Size — small, medium, large are three SKUs.
• Colour or finish — each colourway is its own SKU.
• Pack quantity — a single unit and a pack of six are different SKUs, even of the same item.
• Meaningful spec — a 128GB and a 256GB version are separate SKUs.

What doesn’t need a new SKU is a difference that makes no difference — a minor packaging tweak the customer never chooses between, or a supplier change for an otherwise identical product. Over-splitting creates SKU proliferation and clutter; under-splitting means you can’t tell two genuinely different items apart in a report. Split where the difference is real, and stop there.

SKU examples

A few worked examples show the pattern:

• A red running shoe, size 9: SHO-RUN-RED-09
• A 500ml stainless water bottle: BOT-STL-500
• A large scented candle, vanilla: CAN-VAN-LG

None of these need explaining to someone who knows your system — that’s the point.

How many SKUs should a business have?

There’s no target number — you have as many SKUs as you have distinct products and variants. A single-product seller might have three or four (one per size). A clothing brand can run into the thousands.

The more useful question is whether you have too many. A long tail of SKUs that barely sell — sometimes called SKU proliferation — quietly adds complexity and cost. Reviewing your range with ABC analysis and trimming dead variants (a practice called SKU rationalisation) keeps the catalogue lean and your inventory turnover healthy.

Common SKU mistakes

Most SKU problems are set on the day the scheme is designed, and they are painful to unpick afterwards. The ones that recur:

Encoding something that changes. Supplier, cost, season or storage location inside the code. When any of those change, the SKU either lies or has to be reissued, and reissuing breaks the sales history attached to it.

Reusing a retired code. A discontinued product’s SKU given to a new one merges two products in every report going back years. Retire codes permanently; you will not run out.

Starting a code with a zero. Spreadsheets strip leading zeros silently, which turns 00412 into 412 and creates a duplicate the first time anyone exports a file.

Using characters that cause trouble. Spaces, slashes, hyphens in some systems, and the letters O, I and l, which are indistinguishable from 0 and 1 when someone reads a label aloud or types it from a picking sheet.

Making them too long. Anything past about twelve characters gets abbreviated by whoever is working with it, and the abbreviations diverge.

Letting each channel invent its own. The same product carrying a different code on your website, your marketplace listing and your stock sheet is the most common cause of stock figures that cannot be reconciled.

Creating a new SKU for something that is not a new product. A different carton quantity is usually a new SKU; a new photograph is not. The test is whether you need to count them separately.

Fixing any of these later means a mapping exercise across every system that holds the code, which is why the hour spent on the scheme at the start is the cheapest hour available.

The one-line takeaway

A SKU is your business’s private name for a product — a short, consistent, unique code that lets you track every variant through your system. Build the logic once, keep it consistent, and never confuse your internal SKU with the standardised UPC on the packaging. Get this foundation right and every stock report, count and reorder downstream gets easier.

Frequently asked questions

What does SKU stand for? Stock keeping unit. It’s a unique internal code a business assigns to each distinct product or variant to identify and track it.

What’s the difference between a SKU and a UPC? A SKU is internal — you create it and it only needs to make sense in your business. A UPC is a standardised, universal code (issued through GS1) that’s the same for a product everywhere it’s sold.

Is a barcode a SKU? No. A barcode is the scannable image; a SKU is a number. A barcode can encode a SKU (or a UPC), but they’re not the same thing.

How do I create a SKU? Design a consistent format that leads with the broadest attribute (category), then narrows to variant details (type, colour, size), using short readable abbreviations separated by dashes — and give every variant its own code.

How many SKUs should I have? As many as you have unique products and variants — there’s no set number. Watch instead for too many slow-selling SKUs, and trim the dead ones to keep your catalogue and turnover healthy.

This is the hub for the Stock Identification & Tech pillar. Related guides: what is a barcode · types of barcodes · what is a UPC · what is RFID · how to track inventory.

Written by

↑ Back to top

Warehouse Wise

Warehouse Wise is a free, independent library of guides on inventory management, warehousing and fulfilment, written for the people actually doing the work — new warehouse and inventory staff, small and growing retail and ecommerce businesses, and anyone who holds stock and wants to understand it properly. Every guide explains the term, shows a worked example with real numbers where it helps, and stops. We are independent: we do not sell software or equipment, we take no commission on referrals, and our guides are not ranked by who pays us.