Retail has been one of the slower sectors to adopt spatial computing. While manufacturing, healthcare, and education have moved past the pilot phase with XR technologies, most Australian retailers have treated augmented and mixed reality as novelties — fun for a campaign, hard to justify as permanent infrastructure.

That’s beginning to change. A handful of Australian retailers are deploying spatial computing in ways that go beyond gimmicks, and the early results suggest this technology might finally be finding its footing in physical and digital retail environments.

Product Visualisation: The Clearest Use Case

The most mature application of spatial computing in retail is product visualisation — letting customers see how a product looks in their space before they buy it. This isn’t new. IKEA’s Place app has been doing this since 2017. What’s changed is the quality and the breadth of adoption.

Modern AR visualisation on smartphones and tablets has gotten remarkably good. Lighting estimation, occlusion (virtual objects being hidden behind real objects), and scale accuracy have all improved significantly. When a customer holds up their phone and sees a sofa in their living room, it now looks convincingly real rather than obviously overlaid.

Australian furniture retailers have been early adopters of this technology. Nick Scali launched an AR visualisation feature in late 2025 that covers most of their sofa and dining table range. Early data suggests a measurable reduction in returns — customers who visualise products in their space before purchasing are significantly less likely to send them back because the item didn’t fit or look right.

The economics work. Furniture returns are expensive — shipping a sofa back costs the retailer $200 to $500 per return. If AR visualisation reduces returns by even 10 percent, the technology pays for itself.

In-Store Navigation and Discovery

Larger retail environments — shopping centres, department stores, big-box retailers — are experimenting with AR navigation. The concept is straightforward: point your phone at the space and see directional arrows, store highlights, and product locations overlaid on the camera feed.

Westfield has been trialling AR wayfinding in selected centres, though the current implementation is basic compared to what the technology could deliver. The challenge is less about the AR capability and more about maintaining accurate spatial maps of environments that change frequently — seasonal displays, store refits, pop-up installations.

More interesting is AR-enhanced product discovery. Imagine walking through a wine section and holding up your phone to see ratings, food pairing suggestions, and price comparisons overlaid on the bottles. Or scanning a clothing rack and seeing which items are available in your size, what colours are in stock, and how other customers have styled them.

These applications require robust backend systems — real-time inventory data, product information management, and reliable spatial anchoring. The retailers doing this well have invested as much in data infrastructure as in the AR front end.

Virtual Try-On Gets Serious

Virtual try-on — using AR to see how glasses, makeup, clothing, or accessories look on you — has been around for years in various forms. Most early implementations were gimmicky: low-quality rendering, poor tracking, more entertaining than useful.

The current generation is genuinely good. Apple’s Vision Pro and Meta’s latest headsets demonstrate what high-fidelity try-on looks like with quality hardware. But the bigger market impact is happening on smartphones, where improved face and body tracking lets customers try on sunglasses, lipstick, and jewellery with realistic rendering.

Australian eyewear retailer Oscar Wylee has deployed virtual try-on that captures face shape and size to recommend frame styles before the customer visits a store. The conversion data is encouraging — customers who use the tool before visiting are more likely to purchase and less likely to browse aimlessly.

Where AI and Spatial Computing Intersect

The most promising retail applications combine spatial computing with AI. Computer vision identifies products on shelves and provides contextual information. Recommendation engines analyse what you’re looking at and suggest complementary items. Natural language interfaces let you ask questions about products while viewing them through an AR overlay.

This convergence is where the technology shifts from “nice to have” to genuinely transformative. A spatial computing experience powered by AI can adapt to the individual customer — showing personalised pricing, targeted recommendations, and relevant product information based on purchase history and preferences.

Building these integrated experiences requires expertise across multiple technology domains. Firms like team400.ai work with businesses to connect AI capabilities with spatial computing deployments, ensuring the backend intelligence matches the frontend experience. Without that integration layer, spatial computing in retail remains a visual novelty rather than a revenue-driving tool.

The Challenges Holding Retail Back

Despite the progress, several barriers remain.

Cost of 3D asset creation. Every product in an AR experience needs a 3D model. For retailers with thousands of SKUs, creating and maintaining these is expensive. Automated scanning tools are improving but can’t yet replace manual modelling for high-quality applications.

Customer behaviour change. Most shoppers aren’t in the habit of pulling out their phone for AR while shopping. The experience needs to be genuinely useful, not just available.

Hardware fragmentation. What looks great on a new iPhone looks mediocre on a three-year-old Android. Retailers must decide whether to build for the lowest common denominator or create tiered experiences.

Measurement and attribution. How do you attribute a sale to an AR experience used two days earlier? The measurement frameworks are immature, making continued investment harder to justify.

What Comes Next

The trajectory is clear even if the timeline is uncertain. Spatial computing will become a standard component of retail experiences, not a differentiator. The retailers investing now are building capabilities and customer expectations that will be difficult for competitors to replicate quickly.

The technology is ready. The question is whether Australian retail is willing to invest through the awkward early stages to reach the payoff. Based on what I’m seeing from the early movers, the answer is increasingly yes.