The question is usually asked backwards
“How many models should my range carry?” is the question buyers bring to a supplier, and it is almost always the wrong one. Nobody decides a range by choosing a number and then finding models to fill it. What actually happens is that models are added one at a time, each for a defensible reason, and the number is the residue of those decisions. By the time anyone asks how many there are, the range is already shaped.
The useful version of the question is narrower: which models earn their place, and what does each one cost to keep. A range that is too wide does not fail because of its width. It fails because width is bought with stock, space and attention, and those are finite. Once they run out, the range starts hiding its own best sellers behind its worst ones.
What a model costs once it is in the catalogue
A model that has been added is not a line on a page. It is a commitment across four dimensions, and each one is paid continuously for as long as the model stays in the range:
- Stock. Some quantity has to be held, or the model cannot be sold on the timeline a wholesale customer expects.
- Space. In a warehouse, on a showroom floor, and in the digital sense of a page that has to be maintained, photographed and described.
- Attention. A sales team can explain a short range in detail and can only gesture at a long one, and the gesture is what the customer remembers.
- Obsolescence. Every model that stays eventually has a version that stops being produced, and the last pieces become stock that cannot be replenished.
None of those costs appears when a model is added, and all of them appear later. That asymmetry is why ranges drift wide: adding is a decision made once, and the cost of carrying is a decision nobody makes at all.
Why breadth compounds
Breadth does not add cost linearly, because each model is not a single item. A model in white and a model in a colour finish are two stock positions with different replenishment behaviour. A model offered in a wall-hung and a floor-standing version is two products that share a name. Add finishes and mounting variants and the number of things that have to be forecast, bought, stored and described multiplies while the number of models on the page stays the same.
This is where the per-model minimum on colour and matt finishes becomes a planning fact rather than a footnote. At 30 pieces per model, a colour range is a decision about which models are worth committing to, not a matter of adding a colour to everything. A catalogue that offers every model in every finish is usually a catalogue where several combinations have never actually been ordered.
Three costs that rise with width
The diagram below is the trade-off in its simplest form. It is drawn as three costs against range width, because that is the comparison a buyer actually has to make and it is the one a product list conceals.
The top solid line is stock cover: the more models in the range, the more stock is needed to keep the same depth behind each one, and the harder it is to hold depth at all. The middle line is display space, physical and digital, which grows with every model added and cannot be reclaimed by selling through. The lowest solid line is lead time, which grows because replenishment has to be spread across more items from the same production capacity, so the slowest item sets the rhythm of the reorder.
The dashed line is the one that decides the outcome. It is attention per model, and it falls as width rises. A small team with a narrow range can explain every model and know which customer wants which. The same team with a range twice as wide explains the first few and hands over a catalogue for the rest, and the catalogue does not sell.
Width, depth and cover are one decision, not three
Stock planning treats width, depth and cover as separate variables: how many models, how many of each, and how long the stock has to last. They are not separate. A fixed amount of working capital divides across the range, so every model added reduces the depth available behind the others. That is the mechanism by which a wide range produces stockouts on its best sellers: the money is spread evenly and the demand is not.
Lead time makes the same point from the other side. If replenishment takes 30–40 days from approval, plus transit, then stock cover has to bridge that interval for every model in the range. A wider range with the same total inventory means less cover per model, which means more frequent reorders, which means more of the buyer’s time spent on administration rather than on selling.
A method that produces a range rather than a number
The way to arrive at a defensible range is to rank rather than to count. Four steps, in this order:
- Rank existing sales by model, and look at the tail rather than the head. The tail is where the cost sits.
- Ask what each tail model is for. Some are there to serve a customer segment, some to complete a suite, some because a single account asked once. Each of those is a legitimate reason, and each should be named.
- Decide what a model has to do to stay, in advance, so that removal becomes a rule rather than an argument with a salesperson.
- Decide the entry test for a new model, which is the same question in reverse and is the one most ranges never ask.
The output of that process is a range with a shape: a small group of models that carry most of the volume, a middle group that covers the segments, and a tail that exists for reasons somebody can state. That shape is what a catalogue should express. The count is a by-product.
Range architecture instead of a target count
Three roles are worth distinguishing, because they are managed differently. The models that carry volume need depth and reliable replenishment, and they should never be out of stock. The models that cover a segment need presence rather than depth, and they can be held thinner. The models that exist for a single account or a single project should be treated as project items, ordered against demand rather than stocked, which is possible when the supplier’s minimum allows it.
That last point is where a supplier’s commercial terms change the shape of a buyer’s catalogue. MOQ 0 on white vitreous china means a project item can be ordered without committing to stock, while a colour or matt finish at a per-model minimum of 30 pieces means the same item becomes a stock decision. A buyer who understands that distinction can keep a wide catalogue on paper and a narrow one in the warehouse, which is usually the right answer.
What our range offers a buyer narrowing a catalogue
We make 89 vitreous china models across two-piece, one-piece, wall-hung, back-to-wall, smart, bidet and accessible types. That breadth exists because different markets assemble their ranges differently, not because any one buyer should carry all of it. A buyer building a wholesale range normally takes a handful of families and works within them, and the model pages carry the drawings needed to compare one against another.
The plant is at Chaozhou in Guangdong, built in 2021, 51,000 m², 285 people, with its own tunnel kiln and casting, glazing, firing and QC under one roof. That matters to range decisions in a specific way: a family that is produced in the same plant on the same kiln behaves predictably across its models, so a range built from one family is easier to forecast than a range assembled from several sources.
If you are deciding what your range should carry, send the list you have and the segments you sell to. We will tell you which models overlap each other, which ones share tooling and therefore behave alike, and where a finish turns an item into a stock commitment. Certificate documents on request.
Frequently asked questions
Is a wider range always better for a wholesaler?
No. Width is bought with stock, space and attention, and all three are finite. Past the point where a team can explain and support every model, extra breadth starts to hide the models that actually sell rather than adding choice.
Why does a wide range cause stockouts on best sellers?
Because working capital is divided across the range and demand is not. Each model added takes depth away from the others, so stock ends up spread evenly across models that do not sell evenly, and the fast movers run out while the tail sits.
How does lead time affect the number of models?
A longer replenishment interval needs deeper cover per model, and 30–40 days of production plus transit is a long interval. The longer it is, the fewer models can be carried at a workable depth, which is why lead time belongs in the range decision.
Should colour finishes be offered across the whole range?
Usually not. A colour or matt finish carries a per-model minimum of 30 pieces, so each combination is a stock commitment. Offer colour where the segment asks for it and keep the rest in white, rather than adding a finish to every model on the page.
What is the right way to remove a model?
Set the test in advance, so removal is a rule rather than an argument. Name what each model is for, decide what it has to do to stay, and review against that. Removal decided after the fact always looks like a judgement on somebody’s customer.
Can a catalogue be wider than the stock held?
Yes, where the supplier allows ordering against demand rather than against stock. MOQ 0 on white vitreous china makes that possible for project items, while colour and matt finishes at 30 pieces per model turn the same item into a stock decision.