What counts as one option

Style, colorway, option, SKU

The four words are used interchangeably in conversation and they are not interchangeable in arithmetic. Each one exists because a different function needs to count a different thing, and each has a different owner. Getting them straight is not pedantry — it is the difference between a target you can plan against and a number that has to be re-derived every time someone quotes it.

Style — one design

A style is a single design: one silhouette, one pattern set, one tech pack, one sequence of construction and trim decisions. Everything that makes a style expensive is spent once at this level — pattern development, grading, fit sessions, sample yardage, the factory conversation. That is why style count is the honest measure of development capacity, and why a design team that is asked to hit an option target without a style target has been given an ambiguous instruction. Ten styles in four colorways and forty styles in one colorway are the same option count and completely different workloads.

Colorway — one execution of that design

A colorway is one colour or material execution of a style: the same pattern in a different shade, print, fabric, or finish. Colorways are cheap relative to styles at the development end and expensive at the commitment end, which is the trap. Adding a fourth colorway during a line review costs a lab dip and a swatch on a card; it also adds a full size run to the buy. The asymmetry between how a colorway feels in the room and what it costs in the order is the single most reliable source of range bloat.

Option — the unit the buy commits to

The option is the unit at which a separate commercial decision gets made. In apparel that is almost always the style-colorway, because the colorway is what gets costed, ordered and cancelled independently. In categories where colour is not the multiplier — a furniture model with finish grades beneath it, an accessories body offered in a small material set — the style is the option and colour is an attribute of it. Both readings are in live use, and the gap between them is exactly the average colorway count per style, so a range running three colorways per body reports two numbers a factor of three apart.

SKU — the unit the warehouse picks

A SKU is the smallest individually stocked and counted unit: the option plus size, plus any other dimension that is separately stocked — width in footwear, fill volume in beauty, configuration in furniture. It is the only one of the four units that operations recognises. Every SKU consumes a row in master data, a slot in the warehouse, a line in the allocation run, and a decision in every replenishment cycle after that. The board never shows SKUs, which is precisely why the board understates what it is asking for.

The arithmetic between the units

Two conversions do all the work. Going from styles to options, you sum the colorways behind each style. Going from options to SKUs, you sum the sizes in each option’s run. Both are sums, not products, and the difference between summing and multiplying is where most counting errors are born.

The shortcut form — styles × average colorways, then options × average sizes — is only exact when every style carries the same colorway count and every colorway runs the same sizes. That is almost never true on a real board. One style carries six colorways because it is the hero and three others carry one each; one seasonal print runs a truncated size range because the fabric minimum would not support the full run. The shortcut is fine for sizing a season before the board exists and unreliable for reporting a season that does exist. Use the average to plan the target, use the sum to report the commitment, and never let the two be quoted as the same number.

The second conversion is the one that surprises people. Option count grows linearly with the board — one more card, one more option. SKU count grows with the board multiplied by the size run, so a card added to a category with a long size run costs several times what a card in a short-run category costs, and nothing on the wall communicates that. A board that looks modest at a glance can carry a very large SKU commitment behind it, and by the time that surfaces the range has usually been signed off.

If you want the size-run half of this properly, it is arithmetic that belongs on the plan rather than the board — the size curve calculator on retail-plan.com handles how units split across a run. This page stops at how many things there are.

One board, six cards, three defensible numbers

Here is a six-card board. It is deliberately small and deliberately messy — it contains a pack, a carryover, a made-to-order style, and one colorway that does not run the full size range, because those four things are what make counting hard. Illustrative figures, chosen because they divide cleanly. Not benchmarks, and not drawn from any brand.

Working the SKU column

Card A is four colorways across five sizes: 4 × 5 = 20. Card B is two colorways across four sizes: 2 × 4 = 8. Card C is where the shortcut breaks — two core colorways across six sizes gives 12, the print colorway across three sizes gives 3, so the card is 15, not the 18 a flat six-size assumption would produce. Card D is a pack: two pack assortments in two pack sizes is 4 SKUs, and the individual socks inside the pack are components, not SKUs anyone buys. That is 20 + 8 + 15 + 4 = 47 new SKUs entering the buy from four cards.

Card E is carryover: three colorways across five sizes is 15 SKUs, but they are already live in master data and no new order is placed against them, so they belong in the live SKU count and not in the buy. Card F is made to order: six orderable configurations, zero stocked SKUs, and it occupies a card because a customer can choose it. The board therefore carries 62 live stocked SKUs, 47 of which are a new commitment this season, and 68 things a customer could order.

Note what the shortcut would have said. Eleven bought options at a flat six sizes each is 66 SKUs against an actual 47 — over by nineteen, because socks run two sizes and jackets run four. The average-size shortcut misses in both directions and there is no safe side to err on.

Three counts from the same six cards

Count every style-colorway on the wall and the board holds 15 options: 4 + 2 + 3 + 2 + 3 + 1. Count only what the season is buying — dropping the carryover overshirt and the made-to-order coat — and it holds 11 options. Count the style as the unit and it holds 6 styles, of which five need development work, because the carryover does not. And count only what has been newly developed — dropping the three carryover colorways but keeping the made-to-order coat, which is new this season — and it holds 12 options.

Fifteen, eleven, twelve and six, from one board of six cards. All four are correct, and each is the right input for a different decision: the board and buy option counts for range width against open-to-buy, the newness count for what a returning customer will read as new, the SKU count for master data and warehouse load, and the style count for development capacity. What is not correct is any of them quoted without saying which it is. If you want the season sized before the board exists rather than counted after, the option count planner builds the target up category by category from your own inputs.

Where the count legitimately diverges

Four situations produce a genuine, defensible disagreement rather than a mistake. Each needs a stated rule, because each will otherwise be resolved differently by whoever is building the report that week.

Packs and sets

A three-pack of socks, a gift set, a two-piece swim set sold as one item: one card, one price, one thing the customer chooses. The rule: a pack is one option per pack assortment, and its SKUs are the pack configurations, not the components. The components are consumed by the pack and may also be sold individually, in which case they are separate options in their own right — and if they are, the demand for the component has two sources and the buy has to add them. Counting the components as options inside the pack inflates the range and double-counts the buy.

Size-extended colorways

The same style in the same colour, offered additionally in plus, petite or tall. Some teams count this as one option with a longer size run; others count it as a second option because it is separately graded, separately costed and separately committed. The rule: if it has its own cost, its own buy decision and its own card, it is a second option; if it is a size range on one card with one cost, it is not. Both are defensible; what is not defensible is switching between them mid-season, because the option count then moves without the range moving.

Carryover on the board, not in the buy

A style that returns from a previous season occupies floor space, absorbs customer attention and has to be balanced against the new range — so it belongs on the board — while placing no new order. The rule: keep carryover on the board, flag it visibly on the card, and report the board count and the buy count as two labelled numbers. The failure this prevents is a category whose counts look healthy and whose newness is nil, which stays invisible for exactly as long as carryover is indistinguishable from new on the wall.

Made to order and drop ship

A style a customer can order that the business does not stock: made to order, cut to order, or fulfilled by a partner. It occupies a card because it occupies a choice. The rule: it counts as an option and contributes zero stocked SKUs, which means the card needs a stocked-or-not flag as well as a buy flag. Without that flag the two numbers most likely to be wrong are the ones operations cares about — SKU count and inventory commitment — and they will be wrong upward, which is the direction that quietly buys warehouse capacity nobody needed.

Why your option count disagrees with theirs

Five counting conventions are in common use. They are all internally consistent and they produce different numbers from identical boards. The counts in the third column are the six-card board above, so you can see the spread the choice of convention produces on its own.

The objective is not industry agreement on one convention — it is that your number arrives with its convention attached. Any option target inherited from another team, another season, another vertical or another brand is unusable until the unit and the inclusion rules behind it are stated. A target quoted without them is not a target; it is a number that happens to be the right order of magnitude.

What the card has to say

All of this is solvable at the card level, and it is much cheaper to solve there than in a reconciliation afterwards. A card that carries five fields makes every count in this guide derivable by eye: the style identifier, the colorway count with its swatches, the size run, a buy flag (new buy, carryover, or not bought), and a stocked flag. That is five fields, and four of them are one word or one number.

The size run is the field teams skip, and it is the one that matters most, because it is the only field that converts the board into a SKU number. A board without size runs on the cards cannot be converted to a SKU count without going back to the data, which means it never gets converted during the review — only after it. That is the sequence that produces a signed-off range with a buy nobody sized.

The review question that catches a miscount is one sentence long, and it works because it forces the unit into the open: “how many SKUs is this card?” Ask it of the card someone has just proposed adding. If nobody in the room can answer it, the range is being extended on a number that has not been calculated. Ask it of the card someone has just proposed cutting, too — cutting the long-size-run card and cutting the short one are very different savings, and the board makes them look identical.

The mechanics of laying the board out so these fields are legible are covered in how to build a line board, and the pillar on what a line board is covers where the board sits in the season.

The unit is not the same in every category

An apparel-trained counting instinct transfers badly. The style-colorway-size structure is specific to apparel, and every adjacent vertical either adds a dimension above the style, adds one below the colorway, or removes the size dimension entirely. Each of those changes the ratio between option count and SKU count, which is the ratio every planning assumption is built on.

Apparel

Apparel is the reference case and it still has its own ambiguity, sitting in the size dimension rather than above it. Split runs — petite, tall, plus, short and long inseams — multiply inside what is nominally one option, and the categories that stretch furthest are the ones where sizing is two-dimensional. Bra sizing runs band against cup, so a single colorway can be dozens of SKUs; swim separates are sold as tops and bottoms in different sizes to the same customer; denim runs waist against inseam. These categories break the flat sizes-per-option assumption harder than any other part of apparel, and they are handled specifically in the guide on line boards for denim, intimates and swim.

Footwear

Two things change. The unit sold is a pair, not a unit, so every count that says “units” has to be checked for which one it means before it is compared to anything. And the size run is long — a full adult length run with half sizes is several times an apparel run before widths enter, and adding a second width doubles the SKU count of every option that carries it. An apparel-trained SKU estimate understates a footwear line badly, and the error is multiplicative rather than additive, so it compounds with every colorway approved. It is the clearest case in the cluster for putting the size run on the card. The full treatment is in line boards for footwear and accessories.

Accessories and bags

Most accessories carry no size dimension at all, which collapses the option and the SKU into nearly the same thing. A range of sixty accessory options is roughly sixty SKUs, where sixty apparel options would be several hundred. That is why accessories ranges look small on a SKU report and are not small at all: the work that scales with option count — design, sampling, costing, photography, line-sheet production, and the card itself — is undiminished, while the number operations sees is a fraction of the apparel equivalent. Judging an accessories range by SKU count systematically under-resources it. The exceptions worth flagging are hardware and lining permutations offered as separate orderable items, and belts or gloves that do carry a size run.

Home and furniture

Here the multiplying dimensions are configuration and finish rather than colour and size, and they multiply against each other. A single upholstered model can be offered in several configurations, each in several fabric or finish grades, each grade containing many individual fabrics — so one card is a configurable family, and the number of orderable combinations is a product of the option axes rather than a sum of them. Putting every combination on the wall makes the board unreadable. The workable convention is one card per model, with the axes shown as properties of the card and a count behind each grade, and the SKU question restated as which combinations are stocked versus made to order. That distinction, not the combination count, is what the buy actually turns on. See line boards for home, outdoor and equipment lines.

Health and beauty

The mapping is clean once stated: a shade is a colorway and a size is a fill volume. A complexion product with a shade range in the dozens and a single fill is dozens of options and roughly as many SKUs; a moisturiser in one shade across three fills is one option and three SKUs. The arithmetic breaks on sets and kits, which are heavily used in this vertical. A kit is one option and one SKU, and it is assembled from components that are themselves SKUs with their own demand — so totalling component demand alongside kit demand double-counts, and planning the kit without its component draw starves the individual line. Treat the kit as an option with a bill of materials, not as a bundle of options. See line boards for beauty and personal care.

Sporting goods

Sporting goods adds a level above the style that apparel does not have: the model year. A model carries forward across years with revisions, so the same card can be two options during a changeover, when the outgoing year is still selling through and the incoming year is landing. If the model year is not on the card, the board will either double-count the model or silently drop a year that is still in market, and both errors land in the buy. Below the style, the multiplying dimensions vary by category — team colourways, length or size, and component specification on hardgoods, where two items identical in appearance are different products because of what is fitted to them. The rule is the same one the whole page rests on: put the dimension on the card and the count follows.

Defining the unit is not pricing it

This page defines the unit and counts it. What an option costs to carry — the development spend, the sample and photography load, the inventory it ties up, and the margin it has to earn back — is the economic question sitting immediately behind the counting one, and it is covered in what an option costs on retailnorthstar.ai. The more general question of which level a decision belongs at — style, option, SKU, or something above all three — is planning grain on theapparelos.com. Line board defines and counts the unit; those two price it and place it.

• ✓An option is one commercially distinct thing a customer can choose and a buyer must commit to separately — usually a style-colorway, sometimes the style itself.
• ✓Style, colorway, option and SKU are four different units with four different owners, and each is the right input for a different decision.
• ✓Options are the sum of colorways per style; SKUs are the sum of sizes in each option’s run. Both are sums — the average-based shortcut misses in both directions.
• ✓Packs, size-extended colorways, carryover and made-to-order items each need a stated rule, because each one produces a defensible disagreement rather than a mistake.
• ✓Five counting conventions are in common use and all are internally consistent; the fix is to state the convention next to the number, not to agree on one.
• ✓Put the size run and a buy flag on the card, and ask “how many SKUs is this card?” of anything being added or cut.

• What is a line board? →
• Option count planner (size a season from your own inputs) →
• Option (definition) →
• Option count (definition) →
• SKU (definition) →
• Line boards for footwear and accessories →
• Line boards for denim, intimates and swim →
• Line architecture: good, better, best and price tiers →
• What an option costs (retailnorthstar.ai) →
• Planning grain (theapparelos.com) →

Frequently asked questions

What is the difference between a style and an option?

A style is one design — one silhouette, one pattern, one tech pack. An option is one commercially distinct thing a customer can choose and a buyer commits to separately, which in most apparel usage means one style in one colorway. So a single style offered in four colorways is one style and four options. The distinction matters because the two numbers size different problems: style count sizes development capacity, and option count sizes range breadth.

Is a colorway an option?

In the dominant apparel reading, yes — each colorway of a style is a separate option, because each one is separately chosen, separately costed and separately committed to in the buy. The exception is a category where colour is not the multiplier that matters: a furniture line counted by model with finish grades underneath it, or an accessories carrier counted by body. There the style is the option and colour is an attribute. Neither reading is wrong; a board that does not say which one it uses is.

How many SKUs is one option?

As many as the sizes in that option’s run, plus any other stocked dimension. An option offered in five sizes is five SKUs; the same option offered in five sizes and two widths is ten. There is no general answer, and applying an average size count across a whole range is unreliable in both directions — short runs like socks and one-size accessories pull it down, long footwear runs and two-dimensional bra sizing pull it up. The only accurate method is to sum the actual run behind each option rather than multiply by an average.

What is an option count and what is it used for?

An option count is how many options a range holds, and it is the primary measure of breadth — how wide the range reaches across categories and price tiers, before any question of the depth committed behind each option. It is used to size a season against open-to-buy, to compare categories against each other, and to make over-expansion visible during a line review. It is an internal range-building metric, not a figure presented to wholesale buyers.

Do carryover styles count in the option count?

It depends on which count you are producing, which is why the convention has to be stated. Carryover belongs in the board count, because the customer sees it on the floor and the range has to be balanced including it. It does not belong in the buy count if the season places no new order against it, and it never belongs in the newness count. The workable discipline is to keep carryover on the board, flag it on the card, and report the buy count and the board count as two labelled numbers rather than arguing about one.

Why do two teams count the same range differently?

Because they are using different counting units and different inclusion rules, and usually neither has said so. The unit can be the style or the style-colorway, which alone separates two numbers by the average colorway count per style. On top of that, teams differ on whether carryover, made-to-order and drop-ship items, packs, and size-extended colorways are inside the count. The same board can produce several defensible numbers. The fix is not to agree on one convention across the industry; it is to state the convention next to the number, every time.