When to use a 3D configurator vs a 3D viewer vs just better photography

By Michael O'dlee · April 2026 · 7 min read

A marketing director called me last month with a request I get every few weeks. Their team had budgeted $40,000 for a 3D configurator. Could we build it.

I asked the question I always ask. What is the buyer trying to verify before they commit. There was a pause, then: "We just want the product page to feel more premium." Forty thousand dollars is a lot of money to spend on a feeling. We talked for twenty more minutes and they walked away with a $6,000 photography brief instead. That call probably saved them $34,000 and a maintenance liability they didn't need.

Most products do not need 3D. The ones that do tend to need it for a reason that is specific, measurable, and tied to a single buying decision. Picking the right instrument before you spend the money is the entire game.

This is a guide to picking it.

Most products don't need 3D at all

Before we get to the three options, the honest opener: a lot of brands hear about 3D, watch a Lusion or Active Theory reel, and decide they want some. That is a brand instinct. It is not a buyer-decision instinct.

A good test. If your product is a t-shirt with five colors, a candle in three scents, or a paperback novel, no amount of 3D will change anything. Buyers are not confused. They are not trying to understand internal construction or imagine a custom configuration. They want to see it cleanly and click buy. Spend the money on photography, lighting, and a faster checkout. You will out-convert the brand that built a 3D viewer for the same product.

3D earns its place when one of three things is true. The buyer cannot understand the product from the outside. The buyer needs to assemble a configuration before committing. Or the product only makes sense in operational context. If none of those apply, photography wins.

The three instruments, defined plainly

There are three tools in the toolkit. They do different jobs at different prices. Understanding what each one is actually for, and what it is not for, is the first step.

Better photography

Professionally styled product imagery. Multiple angles, on-figure or in-context shots, lifestyle frames, sometimes a short loop of motion. The classic e-commerce visual layer.

Photography communicates brand, finish, scale, and lifestyle context. It works in every channel without technical overhead. Email, paid ads, print, marketplace listings, the product page itself. One shoot funds a year of marketing assets.

What it cannot do: show internal construction, let the buyer explore at their own pace, or handle deep variant depth without exploding the asset count. A product page with fifteen photos can imply variety. It cannot let the buyer build a configuration.

3D viewer

A single-product 3D presentation. Rotate, zoom, sometimes hotspots or annotations or scripted camera moves. Camera control is the interaction. There is no commerce, no variant building, no order capture.

A viewer is the right tool when the product has form, scale, or internal structure that photography cannot reach. KBIC's facial roller is a good example. The product looks simple from the outside. Its value is in the internal needle array and the rotation mechanism. We built a cross-section view that showed the engineering, and conversion lifted on the product page because buyers could finally see what made the $60 device different from a $20 knockoff. Photography would never have closed that gap.

Zion Dental is a different shape of viewer. A dentist on a chairside iPad walks a nervous patient through an implant procedure. The buyer is making a yes-or-no decision about treatment, not a configuration choice. The viewer's job is to make the procedure legible enough that the patient says yes. That is a single-decision tool. Configurator-class complexity would be wrong.

Dowon APEX is the upper end of viewer territory. A full wastewater-treatment plant in the browser, with twenty pieces of equipment selectable in operational context. Distributors now open sales calls with the walkthrough instead of a slide deck. It replaced the first 40 minutes of pitch. No commerce, no variants, no ordering. Just context that lets a buyer understand a $200,000 piece of industrial equipment without flying to a reference site.

3D configurator

Real-time variant selection. Color, material, size, feature. The 3D scene updates live as the buyer builds the configuration they want. Usually integrated with order capture or commerce.

A configurator is the right tool when the buyer has to assemble a specific configuration before they can commit. Rawfit is the canonical version. Two flagship sneaker models, five colors each, three materials each. On a 2D catalogue the depth disappears. Customers see four photos and assume four options. With the configurator they can pick the exact pair they want, see the leather they chose, capture it, and submit an inquiry. Conversion on configured SKUs ran above unconfigured SKUs in the same category. That is a configurator earning its keep.

Configurators are software. They have ongoing maintenance. New materials, new variants, commerce integrations to keep in sync, browser-version regressions to patch. If the product does not justify that overhead, you have bought a maintenance liability.

Match the tool to the buying decision

The cleanest way to pick is to ignore the tools entirely for a minute and look at the buyer.

What does the buyer need to do before committing.

If the buyer needs to verify what they already want, photography wins. They have done the research. They know the brand. They want a clean look at the thing, and a friction-free path to checkout. Better photography, faster site, cleaner copy. Ship that and stop.

If the buyer needs to understand how the product works, you need a viewer. Internal structure, operational context, multi-audience views. A medical device a patient cannot picture. A piece of industrial equipment that only makes sense at plant scale. A consumer product whose value lives inside the housing. The viewer's job is comprehension, not configuration. The Zion Dental tool does this for patients. The Dowon APEX walkthrough does it for distributors. The KBIC cross-section does it for premium-positioning consumers. Different audiences, same job: make the product legible.

If the buyer needs to assemble a specific configuration before committing, you need a configurator. The variant set is real. The buyer wants to see the exact one they will own. The order data flowing back to your CRM is a separate quiet benefit: you learn what every prospect actually wanted, including the ones who didn't buy.

Three different verbs, three different tools. Verify, understand, assemble.

What each tier actually costs

Honest numbers, based on what we and our peers ship.

Photography. A good product shoot for a single product family runs $3,000 to $8,000. Styling, photographer, retouching, a small motion package. Two to four weeks. The asset library carries a year of marketing.

3D viewer. A single-product viewer with rotation, zoom, and a few hotspots starts around $2,000 and runs to $10,000 for a polished build. One to four weeks of production. Plant-scale viewers like Dowon APEX live in a different bracket because the modeling work is bigger: budget two to four months and $15,000 plus.

3D configurator. A flagship-product configurator with two-to-three-variant axes and order capture runs $15,000 to $50,000 custom-built. Six to twelve weeks. Platform-based configurators (Threekit, Emersya, 3D Cloud) usually cost more over twenty-four months because of per-view fees, but they buy you internal team self-service. The break-even depends on SKU count and how often your team updates content.

The tier matters less than the fit. A $40,000 configurator on a product that needed a $4,000 viewer is a worse outcome than a $4,000 viewer that does its job. Buy what your buyer needs. Not what you wish they needed.

The three mistakes I see most often

Configurator when a viewer would have done the job. This is the most expensive mistake. A brand reads about configurators, picks a vendor, spends $30,000, and ends up with a product where the variant depth is not real. The configurator gets used twice and stops being maintained. The product page is now slower and the marketing team has a ticket queue. A $7,000 viewer would have delivered the brand moment they were actually buying.

Photography when a viewer would have done the job. The opposite case. A brand with a complex product, real internal differentiation, and a buyer who genuinely cannot understand the value from the outside, ships a beautiful photo campaign. Conversions stay flat. The sales team keeps fielding the same five questions on every call. That is the moment a viewer would pay for itself in saved sales hours within a quarter. KBIC's cross-section is exactly this story, told the right way.

Building before knowing the conversion question. The version where the brand picks the tool first, then back-fills a justification. "We're going 3D because our competitor went 3D." That is not a strategy, it is a reaction. The studios who tell you what you want to hear at this stage are not on your side. The studios worth working with will ask what the buyer needs to verify before committing, and they will let you walk away with a smaller answer than you came in with.

The one question to ask yourself

Before you commission anything, sit with this for a few minutes.

What does my buyer need to verify before they commit?

If the answer is "they already know, they just need to see it cleanly," the answer is photography. If it is "they need to understand how this works," you need a viewer. If it is "they need to build their version before they decide," you need a configurator.

That single question will save you more money than any vendor brief, decision matrix, or capability deck. The brand that knows its buyer's verification step is the brand that picks the right instrument the first time. Everyone else spends twice.

If you are stuck on the answer, that is also useful information. It usually means the product positioning needs another pass before the visual layer goes anywhere. Fix that first.

Michael O'dlee is the CEO of CCLEMANG. He leads strategy, client relationships, US operations, and Korean/Japanese market business development.

Back to all insights →