The Future of Product Visualization: Holographic Displays and Beyond
Holographic product displays are not shipping at retail scale. What's shipping today is spatial video and 3D web viewers, built on real, versioned standards: glTF, WebXR, and Apple's visionOS. Glasses-free volumetric holography remains a lab and installation technology, not a checkout-lane one.
What people mean when they say "holographic"
Every few years, a retail trade press cycle declares holographic product displays imminent, usually next to a photo of a museum installation or a trade-show booth. The word covers at least three different technologies, and the gap between them is the gap between what a retailer can ship this quarter and what belongs on a five-year roadmap.
Naming the distinction matters before budgeting for any of it. A vendor pitching "holographic product visualization" could mean a glass pyramid with a hidden screen, a headset app, or a 3D model that spins in a browser tab — three different build costs and three different audiences.
True holography vs. pseudo-holography
True holography reconstructs a light field through diffraction or volumetric projection, so an image carries real depth information from every viewing angle, with no glasses or screen required. The viewer's eyes actually refocus at different depths within the image, the way they would looking at a physical object.
Most retail "holograms" people have actually seen — a floating product spinning inside a glass pyramid — are Pepper's ghost illusions, a reflection trick from the 19th century updated with LED panels and video loops. It looks convincing from a fixed viewing angle and falls apart the moment a viewer walks around it.
Both get marketed under the same word. Only one of them is a light-field reconstruction; the other is a well-executed mirror trick with no genuine depth data behind it.
Spatial video and AR (headset-based)
Headset platforms like Apple's visionOS render stereoscopic 3D content the wearer perceives as having depth, using two offset views rather than a reconstructed light field. This isn't holography by the strict definition — it depends on a device worn on the face — but it delivers a comparable shopping experience today, with a real SDK, real developer tooling, and a real app store behind it.
The distinction is worth keeping precise in any internal pitch deck. Calling a headset app "holographic" invites a comparison to true light-field displays that the headset experience can't actually deliver, and that gap becomes visible the moment a stakeholder asks to see it without putting anything on their face.
3D web viewers (glTF-based)
The least glamorous and most deployed option: a 3D model rendered in a browser, rotatable with a finger or mouse, sometimes with an "AR" button that places it in the room via the phone's camera. No headset, no glasses, works on a phone most shoppers already own.
What's actually production-ready today
glTF: the interchange format doing the real work
Behind every credible 3D product viewer sits glTF, the Khronos Group's royalty-free format for transmitting and loading 3D scenes efficiently. glTF 2.0 is now an ISO/IEC international standard, which means it isn't a single vendor's proprietary format that could disappear with a pivot.
The format handles meshes, materials, textures, and physically based rendering data in one interchange package designed to minimize both file size and runtime processing cost — the two constraints that matter most for a product page that has to load fast on mobile.
model-viewer and the 3D web viewer pattern
Google's open-source <model-viewer> web component wraps glTF rendering and AR placement into a single HTML tag, dropping into a product page without a custom rendering pipeline. It's the pattern most retailers actually ship: one glTF asset per SKU, one web component, and a fallback to a static image where the asset doesn't exist yet.
Production cost: who creates the 3D assets
The bottleneck for most 3D product viewer rollouts isn't the rendering technology — it's producing a glTF asset per SKU. Retailers use either CAD files already produced during product design, exported directly to glTF, or photogrammetry, stitching many photos of a physical sample into a 3D mesh.
CAD-sourced assets are cheaper at scale for retailers who already have design files, since export is close to automatic. Photogrammetry is the fallback for retailers sourcing finished goods with no CAD file, and it's a per-SKU photo shoot, not a one-time tooling investment.
WebXR: shipped, but a Candidate Recommendation, not a Recommendation
The WebXR Device API is the browser-level standard that lets a web page request camera-based AR or headset-based VR sessions. As of its current publication, it sits at Candidate Recommendation status on the W3C's Recommendation track — implemented and shipping, but not yet a finalized Recommendation.
That status matters for planning, not for shipping. WebXR already runs in Chrome and Edge on desktop and mobile, and in Firefox — treat it as stable enough to build on, while expecting incremental spec changes rather than a finished, frozen API.
A Candidate Recommendation is not a warning label. It means the spec is implemented and being tested against real browsers — the same status most of the web platform features retailers already depend on passed through before reaching full Recommendation.
visionOS: windows, volumes, and immersive spaces
Apple's visionOS developer platform organizes spatial apps around three building blocks: windows (2D or volumetric SwiftUI scenes), volumes (3D content in a bounded space), and immersive spaces (content that isn't bounded to a window at all). A retailer building a flagship visionOS shopping experience is choosing among these three, not inventing a new interaction model from scratch.
This is real, shipping developer tooling with sample projects and a production app store behind it. It is also a single-vendor, single-headset platform — a legitimate flagship channel, not yet a mass-market one.
What's still a lab technology
True volumetric and light-field holography
Glasses-free displays that reconstruct a real light field, viewable from any angle without a headset, remain research-stage and niche-installation technology — trade show booths, science museums, specialized signage. No standards body publishes a production-scale retail deployment claim for this category, and neither does this article.
Budgeting for a glasses-free holographic kiosk as a near-term retail channel is planning against a technology that hasn't reached the standardization stage the web and headset technologies above already have.
Haptic feedback
Simulating the feel of a product's texture through ultrasonic pulses or vibration is an active research area with genuine promise for high-consideration categories like apparel and furniture. It has no broadly shipped, standardized API comparable to glTF or WebXR, which means any deployment today is a custom, vendor-specific integration.
Treat any haptic pilot as a research spend with a specific learning goal, not as a feature on a product roadmap with a ship date. The tooling isn't standardized enough yet to commit engineering time against a specific vendor's API surviving unchanged.
A vendor evaluation checklist
Before signing with any "holographic" or "3D visualization" vendor, run their pitch against a short list of concrete questions.
- Does the output use glTF as the underlying asset format, or a proprietary format that locks the catalog to one vendor?
- Does the experience degrade to a static image or basic 3D viewer on devices that don't support AR or headsets?
- Who owns the 3D asset pipeline going forward — the vendor, or an in-house team producing new SKUs weekly?
- Is "holographic" in the pitch a true light-field display, a Pepper's ghost illusion, or a headset app — ask directly?
- What's the per-SKU cost and turnaround time to add a new product to the catalog, at the retailer's actual SKU velocity?
The standards actually doing the work
| Standard | What it standardizes | Status | Who ships it |
|---|---|---|---|
| glTF 2.0 | 3D asset format — meshes, materials, PBR data | ISO/IEC international standard | Every major 3D tool and browser viewer |
| WebXR Device API | Browser API for requesting AR/VR sessions | W3C Candidate Recommendation | Chrome, Edge (desktop + mobile), Firefox |
| visionOS (windows/volumes/spaces) | Native spatial app architecture | Apple platform SDK, actively developed | Apple Vision Pro only |
| Volumetric / light-field holography | Not yet standardized at a body like W3C or Khronos | Research and niche installation | Specialized vendors, no common API |
Where this fits a retail roadmap today
The honest sequencing runs from cheapest and most broadly reachable to most expensive and most narrowly reachable. A 3D web viewer reaches every shopper with a modern phone browser.
A WebXR "view in your space" feature reaches a subset of those phones. A headset app reaches only shoppers who own that specific headset.
- Invest first in glTF-based 3D product viewers for high-consideration SKUs — furniture, footwear, anything shoppers rotate mentally before buying
- Add a WebXR "view in your space" fallback for phones that support it, degrading gracefully to the 3D viewer where it doesn't
- Treat visionOS or any headset-based experience as a flagship or pilot channel, not a core conversion path, until headset adoption clears mass-market thresholds
- Don't budget for glasses-free holographic kiosks as a near-term line item — track the category, don't build against it yet
Ship the glTF pipeline before the headset app. The 3D viewer reaches every shopper with a phone; the headset experience reaches shoppers who own a specific device that hasn't reached mass-market volume.
Failure modes and pitfalls
Chasing the demo over the pipeline
A single hero product with a beautiful 3D model is a demo. A retailer with 5,000 SKUs needs a repeatable, semi-automated 3D asset production pipeline, or the "innovative" product page becomes three products deep and never expands.
No fallback for non-AR devices
Not every shopper's phone or browser supports WebXR. A product page that assumes AR support without a static-image or 3D-viewer fallback loses those shoppers entirely instead of degrading gracefully.
Ignoring accessibility in immersive experiences
Motion in a 3D viewer or immersive space can trigger discomfort for some users, and screen-reader support for spatial content is far behind 2D web accessibility tooling. Provide a static, keyboard-navigable alternative view for every 3D experience, not just a loading spinner.
FAQ
Is holographic product display ready for retail today?
Glasses-free, true holography is not. 3D web viewers built on glTF, and headset-based spatial shopping on platforms like visionOS, are real and shippable today.
What format should a retailer standardize product 3D assets on?
glTF 2.0. It's an open, ISO-standardized format supported by every major 3D authoring tool and web viewer, which avoids locking a catalog into one vendor's proprietary format.
Does WebXR work on iPhone Safari?
Check current browser support directly before committing — WebXR's implementation status varies by platform and changes as the spec moves through the W3C process. Build a non-WebXR fallback regardless.
Is Apple Vision Pro worth building for?
As a flagship or pilot experience for a premium or design-forward brand, yes. As a primary conversion channel, not yet — it's a single-device platform with real developer tooling but limited market reach today.
What's the actual retail benefit of 3D product viewers?
They let a shopper inspect a product from angles a photo set can't easily cover, particularly for furniture, footwear, and anything with meaningful physical dimension. Measuring the conversion impact requires a retailer's own A/B test — no general figure applies across categories.
Should we build a custom 3D viewer or use an off-the-shelf component?
Start with an existing open-source component like <model-viewer> rather than a custom renderer. The engineering cost of a custom WebGL or WebGPU pipeline rarely pays for itself against what an established, actively maintained component already handles.