Retail has been one of XR’s most promising application areas for years now, and honestly, the reality in 2026 is more nuanced than the hype suggested. Some things have absolutely taken off. Others are still struggling to justify themselves. The interesting question isn’t whether XR works for retail — it clearly does in some contexts — but which deployments are actually delivering ROI and which are still expensive experiments.
Virtual Try-On Has Become Mainstream (for Some Categories)
If you’ve used the Snapchat camera to try on sunglasses, tested lipstick shades on Sephora’s app, or seen how a sofa fits in your living room through IKEA Place, you’ve already used consumer AR without necessarily thinking of it as XR. This category has genuinely matured.
Fashion accessories — eyewear, jewellery, hats, trainers — translate particularly well to AR try-on because the spatial accuracy requirements are achievable with phone cameras, and the purchase decision is heavily influenced by how something looks. Warby Parker’s virtual try-on has been running for years and they’re open about the fact that customers who use it are more likely to buy and less likely to return. Specsavers rolled out AR try-on across their UK web presence in 2025 and reported a measurable drop in return rates.
Clothing is harder. Fabric drape, fit across body proportions, and how garments move are genuinely difficult to simulate accurately with current phone-based AR. Platforms like Zara and ASOS have piloted clothing try-on, but the experience is noticeably less convincing than eyewear. For clothing, the more honest use case is size recommendation — using body measurements (phone scan, or data from past purchases) to suggest sizing rather than rendering the garment in 3D on your body.
Home furnishings sit somewhere in the middle. The spatial accuracy of phone LiDAR (standard on high-end iPhones and now several Android flagships) is good enough that placing a sofa or bookshelf and walking around it feels genuinely useful. The gap is in material rendering — fabric textures under variable home lighting still look slightly off compared to reality.
In-Store XR: Where It’s Actually Worth Deploying
Here’s the thing about in-store AR: the ROI calculation looks very different depending on what you’re selling. For high-consideration, high-margin categories — kitchen design, bedroom furniture, fitted wardrobes, car customisation — in-store AR has become a legitimate sales tool.
Wickes and Howdens have both deployed in-store kitchen planning tools with AR visualisation. A customer walks in, discusses their kitchen dimensions and style preferences, and leaves with a photorealistic walkthrough of their planned kitchen rendered in 3D. The conversion from planning session to sale is substantially higher than with flat 2D designs, partly because the customer feels they’ve already “seen” the kitchen.
Car dealerships have been running AR configurators for a few years now, and the adoption has been consistent enough that it’s no longer a differentiator — it’s expected. You can configure colours, wheel options, and interior trim on an AR model in the showroom or at home before your test drive. Manufacturers like BMW, Mercedes, and Jaguar Land Rover have their own apps; dealership groups have third-party solutions.
For fast fashion, electronics, and everyday groceries, in-store AR makes less sense. The purchase decision is low-consideration, margin is tighter, and there’s no AR use case that justifiably pays for itself.
The Headset Question
Most successful retail XR in 2026 is phone-based, not headset-based. This is a practical reality: you can’t ask customers to put on a headset in a shop without significant friction. Staff-facing headset applications are different — enterprise AR for stock management, planogram compliance checking, and guided receiving are genuine headset use cases where the headset is a tool for employees, not customers.
Zara ran a brief trial with shared headsets for in-store product visualisation. The hygiene concerns (sharing headsets between customers in a public retail environment), the staff time required to manage them, and the customer reluctance to wear unfamiliar devices all combined to make it operationally difficult. They’ve since pivoted to phone-based experiences.
The exception is destination retail and luxury. High-end car showrooms, bespoke kitchen studios, and luxury jewellers can justify dedicated iPad or headset experiences because the customer is already there for a considered, relationship-driven purchase. Louis Vuitton’s flagship stores have experimented with headset-based product heritage experiences — not quite try-on, but immersive brand storytelling — with positive reception from customers who choose to engage with it.
What the Data Actually Shows
Conversion rates and return rates are the two metrics retailers care about most. The evidence on both is reasonably consistent: AR try-on increases conversion for categories where spatial accuracy is achievable, and reduces returns for the same categories. The effect sizes vary but are generally in the range of 10-25% improvement, which at retail scale is significant.
The honest caveat is that most of this data comes from vendors selling AR solutions, and independent audits are rarer. The category leaders — IKEA, Specsavers, Warby Parker — have been transparent enough that their results are credible. Treating vendor case studies from newer entrants with a bit more scepticism is fair.
There’s also the engagement versus conversion question. AR try-on drives engagement metrics reliably. Whether that engagement converts to purchases depends heavily on how well the AR experience integrates with the checkout flow. A beautiful AR experience that requires four taps to get from “I like this” to “I’ve ordered it” loses conversion at each step.
Where Things Are Heading
Phone camera quality continues to improve, which directly benefits AR try-on fidelity. Neural rendering techniques — using generative models to produce photorealistic material previews rather than purely geometric simulation — are starting to appear in production tools and will improve clothing and fabric AR substantially over the next couple of years.
The smart glasses category is also evolving. If Ray-Ban Meta or the upcoming Google/Samsung Android XR smart glasses reach meaningful consumer adoption (still a big if), passively-triggered AR shopping experiences become possible — walking past a shop window and getting a prompt to try something on in context. That’s a qualitatively different use case from pulling out your phone and opening an app. Whether consumers will want it is a separate question from whether the technology can do it.
For now, the practical advice for retailers is: virtual try-on works for accessories, home products, and car customisation. Start there, measure carefully, and build the business case before expanding into harder categories.