One Model or a Range? Choosing Your First Assortment

The one-model start is underrated

Sellers beginning in this category tend to over-assort: three or four models chosen to cover imagined customer types, none of them ordered deeply enough to earn real pricing, all of them splitting attention. The alternative is underrated: one well-chosen model, ordered at a quantity that earns terms, sold with listing copy written once and properly. The first assortment decision is not which models; it is how many.

When one model is the right answer

Start with one when the channel is narrow, the capital is limited, or the category is unproven for your audience. One model concentrates every advantage a small seller has: the sample approval is thorough, the listing is excellent because it had one job, support answers are consistent because they have one subject, and the reorder carries data instead of hope. Choose the model the way a store builds a ladder's middle step: the mid price point, the documented band list, the display that passes the daylight test, and the supplier who documents spares.

The signals that call for a second model

Add a model when the data asks for it, not when the calendar does. The signals: customers asking for a feature the current model lacks, often enough to count; a price-point request from below or above that recurs; and sell-through fast enough that a second SKU shares the reorder freight rather than splitting it. The second model should differ visibly, in display quality, mounting type, or the GPS build, because two similar models split the same sale and double the support.

Building the ladder deliberately

The mature assortment is the three-step ladder from the stores: entry, volume mid, and one feature model at the top, differentiated by what a customer can see. Each step earns its slot with its own sell-through and its own support cost, reviewed quarterly. The ladder grows one rung at a time, and each addition is a small business case: expected volume at the new price point, the support it adds, and the spares it obligates.

The costs a range carries that one model does not

• Listing and photography per SKU, done properly.
• Sample approval per model, not skipped because the range "has one already."
• Spare parts per model, for the life of the installed base.
• Support knowledge per model, in every staff member's head.

The costs are the reason the one-model start keeps winning for small sellers: every cost on the list scales with the count, and the count is the only number fully under your control.

The honest test

Before adding any model, write one sentence: which customer, asking for what, does this model serve that the range does not. A sentence that answers with data is a model worth sampling. A sentence that answers with an adjective is a shelf slot waiting to be discounted at the end of its first season. Assortment is an opinion about customers; the data, not the catalog, writes the opinion.

The range's memory page

Whatever the assortment's size, keep one page of memory per model: why it was added, what it was expected to do, what it actually did at each review, and what its retirement condition would be. The page takes five minutes per quarter per model and turns the range from a collection of SKUs into a portfolio with reasons. New staff read it to understand the shelf; suppliers hear it in sharper questions; and the assortment meeting, where growth is argued, starts from evidence instead of enthusiasm. The memory page is the smallest possible strategy function, and the range that keeps one never grows by accident.

The one-model year, run properly

A single-model year is not a small version of a range year; it has its own playbook. The depth goes into one listing that can outperform any range page: full photographs, the tuning section, the honest range paragraph, the box-contents layout shot. The support desk learns one unit completely, and the macros become genuinely useful. The buying concentrates: sample approved once, reorder timed on real sell-through, terms renegotiated from evidence at the year's midpoint. Sellers who run the one-model year this way end it with three assets competitors cannot shortcut: a proven SKU, a documented supplier relationship and a customer list that trusts the store's next decision.

Reading the moment to widen

The moment to widen announces itself in three signals at once, not one: recurring requests for a specific difference, sell-through deep enough that a second SKU shares freight, and support questions the current model structurally cannot answer. All three, because each alone has a cheaper response: a request alone is a counter answer, depth alone is a bigger reorder, support questions alone are a listing edit. The seller who waits for the trio widens once, deliberately, and the second model starts life with the first model's audience already primed.

Keeping the opinion as the range grows

As the assortment widens, the original opinion, about which customers the store serves, needs a guardian. The quarterly review keeps it: each model's slot restated in customer terms, retirements made on the arithmetic, additions justified by the trio. The guardian is a document, one page, updated at the review, that any new staff member can read in five minutes and understand what the store sells and why. Ranges drift into warehouses when nobody holds the opinion; ranges with a guardian stay catalogs, and catalogs are what customers return to.

The second model's first month

A new model's first month decides its reputation faster than its first quarter, so the widening plan includes the launch care: the listing complete before stock lands, the support macros written from the sample check, the counter card printed, and the first fifty customers receiving the setup follow-up the first model's buyers learned to expect. The second model inherits an audience that compares it, fairly or not, to the first model's experience; the launch care is how the comparison starts in the new model's favor. Widening is a supply decision; launching is a promise, and the month it is kept is the month the range earned its second rung.