3D product configurator: how it is built

Twelve days, and what that figure really says

Twelve days passed between the first meeting with the shipyard and going into production. The tool lets a buyer compose their boat, hull, superstructure, deck finishes, upholstery, engine package and electronics, and produces a priced specification sheet.

That timeline surprises people, and it is often misread. It does not mean building a configurator is trivial. It means the sales director had the catalogue and the compatibility rules in his head, and agreed to spend a day or two on them at the outset.

On projects where those days do not happen, the same tool takes considerably longer. The gap is not in the development, which takes roughly the same time either way. It is in work nobody had identified as work.

The rules nobody ever wrote down

In any company selling a configurable product, there is a set of rules saying what goes with what. This engine package requires that electrical pack. This finish does not exist on that model. Above a certain dimension, a reinforcement is required.

Those rules are perfectly well known. They are simply stored in two or three people's heads, in the workshop and the design office, and applied when a quote is reviewed. Nobody wrote them down, because nobody ever needed them in writing.

A configurator does need them. It has to refuse a combination before a buyer composes it, not after a salesperson reviews it. That is why the first session of a configurator project is almost always the first time those rules are said out loud.

That session has a side effect clients often mention before any other benefit: it surfaces internal disagreements. Two people state the same rule differently, or a salesperson has been offering an option the workshop considers discontinued. Those gaps are settled once, and they stop coming back.

This is why the schedule of a configurator project does not read from the number of screens. It reads from the state of the catalogue and the availability of whoever can make the calls.

What gets built, and in what order

On the projects we run, the build order is almost always the same, and each layer depends on the one before it.

• The option catalogue: families, variants, references, prices. Catalogue work, not development, and it sets the go-live date.
• The compatibility rules: what excludes, what requires, what depends on a dimension.
• The rendering: producing the visuals and assembling them live.
• The output: the summary, the price, the specification sheet and its delivery.

Projects that slip start with the rendering, because that is the part you can show in a meeting. Projects that hold their schedule start with the catalogue, even if the first demo runs on placeholder images.

A configurator with beautiful images and approximate rules produces quotes the workshop rejects. The reverse, placeholder images and correct rules, produces signable quotes in the first week.

The visuals, briefly

Two families of answers exist. The first composes images in layers: one base visual per viewpoint, then one layer per variant, stacked according to the buyer's choices. The second displays a 3D model the buyer can rotate freely.

In both cases the visuals are produced once and reused by every combination. You never photograph every possible combination: you photograph each variant once.

The trade-off between them, with the calculation that settles it: 2D or 3D configurator, which one for your product

The products where it earns its place

A configurator pays off when three conditions meet: the product comes in visible options, unit value is high, and the sales cycle involves back and forth.

Marine, leisure vehicles, furniture and joinery, machinery and industrial equipment, interior fit-out: these are the families where those three conditions most often come together.

The right indicator is not the number of catalogue references. It is the number of combinations your sales team prices by hand each month, because that is exactly the volume a configurator takes over.

Conversely, a product sold in two variants and shipped the next day does not need one. A good product page is enough, and costs ten times less.

Budget and timelines, by the state of your catalogue

The schedule follows directly from the state your catalogue is in.

• One to two days to write the catalogue and the rules, when the information already exists in the company.
• One to two weeks to build the configurator itself.
• The rest of the calendar depends solely on producing the visuals, when they do not yet exist.

On budget, a configurator sits on average between 6,000 and 25,000 euros. That average has no limit at either end: a single range with existing visuals costs less, a full catalogue connected to pricing and production costs more. The gaps almost always come from the number of visual variants, rarely from the code.

One line is worth negotiating before signature rather than after: who updates the catalogue. On our projects, new models and prices are entered by the client, with no developer involved. That is the difference between a tool that lives and one that goes stale in six months.

A budget broken down line by line: what custom software costs

What changes once the tool is in service

The shipyard configurator now produces specification sheets precise enough to be quoted without a single clarifying email.

The first effect is a shift in the sales conversation. The buyer arrives having already composed, compared and hesitated. The meeting therefore starts much further along and is about the decision, not the description.

The second is the disappearance of an entire category of errors. When impossible combinations are never offered, they stop reaching production.

The third is the one nobody asked for, and often the most appreciated: sales, the workshop and the customer finally look at the same configuration, with the same options and the same price. Order misunderstandings, which are expensive and surface late, disappear with it.

If you are considering a project like this, the first question to ask yourself is not technical: who, at your company, could spend a day or two writing down your catalogue's rules? The answer determines your schedule more surely than any specification.