What a 3D Product Page Actually Costs (and What It Should Return)
The honest cost-and-ROI breakdown for a 3D product page in 2026. Cost bands by capture method, lift bands by category, and a worksheet a CFO will sign off on.
If you run finance, growth, or merchandising at a premium DTC brand, the question keeps coming back. Should we put a 3D viewer on the product page? What does it cost? What does it return? The vendor decks all pitch the upside. The honest math is simpler, and it works for or against you depending on the category and AOV. Here is the cost side, the lift side, and a worksheet you can fill in for your own catalog.
What a 3D product page costs and returns
A 3D product page costs $50 to $500 per SKU on the capture side, plus a flat monthly app fee for the viewer software. The lift is concentrated in a few categories. Shopify reports Rebecca Minkoff buyers who interacted with 3D models were 27% more likely to place an order, and Gunner Kennels saw a 40% increase in order conversions and a 5-point drop in returns after AR and 3D shipped on the page. At a 25% conversion lift on a $300 AOV product with 5,000 monthly visits and a 2% baseline conversion rate, a 3D viewer adds about 25 conversions a month, or $7,500 in incremental revenue. Most premium DTC stores break even in week one.
That is the headline. The rest of this piece is how to size it for your own SKUs.
The cost side: three capture methods
The cost of putting a 3D version of a product on the page comes from two lines. Capture (turning the real product into a 3D file) and software (the viewer that shows the file on the page). Capture is the variable line. Three methods, three cost bands.
Hand-modeled GLB. A 3D artist sculpts the product from scratch using reference photos and dimensions, then tunes PBR materials. This is the gold standard for products with complex geometry that need to be configured (sofas, chairs, lamps). Cost runs $250 to $500 per SKU on a freelance budget, more on an agency budget. Turnaround is one to three weeks per product.
Photogrammetry. A capture rig or a photographer shoots the product from 50 to 200 angles, then software stitches the photos into a textured mesh. The result is photoreal but heavy, often 20 to 80 MB per file before compression. Cost runs $150 to $400 per SKU. Turnaround is a few days.
Gaussian splat. A new method that turns 2 to 8 product photos into a photoreal 3D scene a buyer can rotate. The output is small (1 to 4 MB after compression), photoreal, and fast to produce. Cost runs $50 to $150 per SKU. Turnaround is hours, not days. The Gaussian splatting for ecommerce guide covers what each method actually produces and which buyer questions it answers.
For most premium DTC stores starting today, the math points at Gaussian splat first. You can put a viewer on a hero product for under $200 and find out if the lift is real before you commit to a full catalog rollout.
The software side: app fee or custom build
Two paths. Subscribe to a viewer app or build a custom Three.js component.
Shopify or platform app. Most 3D viewer apps on Shopify charge $30 to $300 a month for unlimited SKUs. The fee is flat, not per-product. The 3D product viewer comparison walks through which one fits which buyer flow.
Custom Three.js build. A custom viewer built on Three.js or model-viewer gives you full control over how the rest of the page reacts to what the buyer does inside the viewer. The cost is a one-time development charge, usually $5,000 to $25,000 depending on whether you ship a basic spin-and-zoom or a full configurator. The Three.js product viewer tutorial covers what a shippable viewer needs.
For a premium DTC brand testing the waters, an app subscription wins. You ship the test in a week, prove the lift, and then decide whether to invest in a custom build for year two.
The lift side: what 3D actually returns
This is the part vendor decks oversell. Real lift varies by category, AOV, and how the viewer is integrated with the rest of the page. The honest bands, by published case data:
Conversion lift. Across published Shopify case studies, the cluster sits between 20% and 40% on the SKUs where 3D ships. Shopify reports Rebecca Minkoff buyers who interacted with 3D models were 44% more likely to add to cart and 27% more likely to place an order. Customers who viewed products in AR were 65% more likely to purchase. Gunner Kennels saw a 40% increase in order conversions. Shopify's furniture write-up cites Houzz research showing consumers are 11 times more likely to buy furniture after viewing it through AR, and notes a survey finding that 80% of consumers who used AR to shop said it increased their purchasing confidence.
Return reduction. The same Shopify Gunner Kennels case shows a 5-point drop in returns after AR and 3D shipped. 58% of consumers who used AR believe the technology prevented them from needing to return a purchase. The returns reduction with 3D and AR breakdown walks the math by category.
General industry lift. Customers using AR spent around 20% more time browsing and viewed 1.28 times more products, with purchase likelihood up almost 20%.
For a CFO model, use 25% as the conservative conversion lift assumption on the SKUs where 3D ships in fitting categories (furniture, jewelry, watches, eyewear). Use 0% lift on apparel, footwear, and headphones. Use a 3-point return reduction on furniture and 1 to 2 points on the other fitting categories.
The ROI math: a worksheet a CFO will sign off on
The formula is short. Plug in your own numbers.
Monthly incremental revenue =
Monthly visits to SKU
× Baseline conversion rate
× Conversion lift from 3D
× Average order value
Then subtract:
Monthly cost = (Capture cost / amortization months) + Monthly software fee
Net = monthly incremental revenue minus monthly cost.
Worked example. A premium furniture brand at $1,200 AOV ships a Gaussian splat on its hero sofa. The page gets 5,000 visits a month with a 2% baseline conversion rate. Capture costs $150, amortized over 12 months. Software costs $99 a month.
- Monthly conversions today: 5,000 × 2% = 100
- With 25% lift: 5,000 × 2% × 1.25 = 125 (25 extra conversions)
- Incremental revenue: 25 × $1,200 = $30,000 a month
- Monthly cost: ($150 / 12) + $99 = $111
- Net: $29,889 a month
The break-even on capture cost is roughly the first six visits. The page is in the black on day one.
Second example, smaller catalog. A jewelry brand at $300 AOV with 3,000 monthly visits and a 1.5% baseline. Capture $80, software $49 a month.
- Conversions today: 3,000 × 1.5% = 45
- With 25% lift: 56 (11 extra)
- Incremental: 11 × $300 = $3,300 a month
- Monthly cost: ($80 / 12) + $49 = $56
- Net: $3,244 a month
The math holds even when AOV and traffic are modest, as long as the category is one where the buyer's question is geometric.
Where the math breaks down
Three categories where 3D does not earn its keep, even at zero capture cost.
Apparel. The buyer's real question is fit and feel. A rotating mannequin does not answer "does this shirt feel scratchy" or "does the waistband sit right." The AR product try-on category breakdown walks through which apparel sub-categories see any lift (almost none, outside virtual try-on for very specific sub-segments).
Footwear. Same problem as apparel, plus the geometry of a foot inside a shoe. AR try-on shows silhouette but not fit. Most footwear brands see no measurable lift.
Headphones and IEMs. The buyer wants to know how they sound, how they fit over 8 hours, and whether the soundstage matches their genre. A viewer answers none of those. The audio equipment product page guide covers what does work.
If you sell in those three categories, do not try to force a positive ROI on a 3D viewer. Spend the budget on size guides, fit data from past buyers, and independent audio reviews. The lift will be larger and cheaper.
The hidden costs most calculators miss
Five lines that vendor decks usually leave out.
Re-shoots when the product changes. A new colorway, a new fabric, a small geometry change all need a fresh capture. Plan for a re-shoot every time the SKU's photo set changes. Gaussian splat from existing photos keeps this cost down, since the photos for the new colorway already exist.
Mobile bandwidth on LTE. A 12 MB GLB does not load before the buyer leaves. The Gaussian splat mobile performance breakdown covers the file size targets that survive a coffee-shop connection. If the viewer adds two seconds to mobile load, you lose the conversion lift to a slower page.
CDN hosting fees. Most 3D viewer apps include CDN delivery in the monthly fee. Custom builds need a separate line for CDN bandwidth, usually $5 to $50 a month per hero SKU at premium DTC traffic.
Staff time to script the tour. A spin-and-zoom is the floor. The lift comes from a guided tour that points the buyer at the materials, the dimensions, and the back of the product. Plan for 1 to 3 hours of merchandiser time per SKU to script the camera path and the on-model labels. The furniture configurator guide covers what a useful tour looks like for room-scale categories.
Page weight tradeoffs. Adding a 3D viewer to a page that already runs hot on Core Web Vitals can tank LCP. Test on real mobile before you commit. The Shopify PDP conversion checklist walks through how to keep the rest of the page light.
The 30-day test plan
The cheapest way to find out if 3D earns its keep is to test one SKU.
Day 0 baseline. Pick the SKU where the buyer asks the most questions or where the return rate is the highest. Pull 30 days of conversion rate, AOV, and return rate by reason code.
Day 0 capture and ship. Use the capture method that fits your budget. For most brands, a Gaussian splat from existing product photos is the fastest test. Ship the viewer on the page. Do not change anything else on the SKU during the test window.
Day 1 to 30 measurement. Track conversion rate and the "didn't match expectations" return reason code. Compare to the 30-day baseline.
Day 30 decision. If conversion lift clears 10% on a fitting category SKU, roll the same workflow to the next four hero products. If lift is under 5%, the test product was not the right fit. Try a different SKU before scaling. If lift is negative on mobile but positive on desktop, your file size is the problem, not the 3D itself.
The discipline that matters: do not roll across the catalog before the test ships clean numbers. A bad first ship taints the rollout for the next year.
Where Oeave fits
Oeave handles the capture-cost line. Two to eight product photos in, a Gaussian splat on the page out, no 3D artist, no theme rebuild. The capture cost falls into the low end of the band ($50 to $150 per SKU). The viewer ships on top of your existing product page through a small embed, not a theme rebuild. For brands that have not tested 3D yet because the cost-per-SKU math felt risky, this is the test path with the smallest downside. Pick one SKU, run the 30 days, decide from numbers.