Walk into certain Westfield locations in Sydney and Melbourne these days, and you might notice people holding up their phones, following virtual arrows overlaid on the real world. AR wayfinding has moved from airport trials to retail deployment, and the results are… mixed.

What’s Actually Being Used

The current crop of retail AR wayfinding breaks down into two camps: the app-based systems and the WebAR approach that works through a browser.

App-based systems—the kind you’ll find at some larger shopping centres—require downloading software that’s usually 50-100MB. They work well once installed, with solid tracking and the ability to remember your parking spot. But the conversion rate is brutal. Most shoppers won’t download an app for a single visit.

WebAR systems dodge that problem entirely. Scan a QR code, grant camera access, and you’re navigating. The tracking isn’t quite as stable, and features are more limited, but actual usage is reportedly 10x higher than the app-based alternative.

Where It Works

Large, multi-level shopping centres are the obvious use case. Finding a specific store in a sprawling complex like Chadstone or Westfield Bondi Junction can be genuinely annoying, especially if you’re trying to locate something like a customer service desk or parent’s room.

The AR overlay showing you which escalator to take, or that your destination is one floor up and 50 metres ahead, solves a real problem. Anecdotally, centre management reports fewer “where is X?” queries at information desks.

Hardware stores are another sweet spot. Try finding a specific type of screw in a Bunnings, and you’ll appreciate why AR navigation in large format retail makes sense. Some specialists in this space are working on systems that don’t just show you where to go, but integrate with inventory data to confirm the item’s actually in stock at that location.

Where It Doesn’t

Small to medium retail? Questionable value. If I can see the entire store from the entrance, I don’t need AR to find the denim section.

The tech also struggles in environments with lots of glass and reflective surfaces—which describes most modern retail architecture. The computer vision systems can get confused, leading to directional arrows that drift or point the wrong way. Nothing kills user trust faster than following AR directions into a wall.

Battery drain is still an issue, particularly on older phones. Running AR tracking, camera, and GPS simultaneously will hammer your battery. If you’re already at 30% after a day out, firing up AR wayfinding might not seem worth it.

The Real Barrier: Setup Cost

The bigger challenge isn’t the customer-facing tech—it’s the backend infrastructure required to make it work.

Retail AR wayfinding needs a detailed 3D map of the space, with precise positioning of every store, facility, and landmark. Creating this usually involves a professional survey using specialized equipment, plus ongoing updates whenever the floor plan changes.

For a major shopping centre with dozens of tenants constantly rotating, that’s significant overhead. Some centres are exploring cheaper alternatives using Google’s ARCore and crowdsourced mapping data, but accuracy suffers.

Then there’s the positioning technology itself. Most systems combine GPS (which doesn’t work indoors), WiFi triangulation, Bluetooth beacons, and visual positioning. That means installing and maintaining beacon infrastructure throughout the venue—more capex, more ongoing maintenance.

The Ikea Approach

Ikea’s AR app remains one of the better implementations, though it’s solving a different problem. Their Place app lets you visualize furniture in your home before buying—not navigation, but spatial computing doing something genuinely useful.

The conversion impact is real. Being able to see whether that bookshelf actually fits your space, and whether the colour works with your existing furniture, reduces returns and increases purchase confidence. That’s AR solving a problem that couldn’t be solved any other way.

What’s Next

The next wave will likely be more integration with retail apps people already have installed. If you’re using the Myer or David Jones app for click-and-collect, adding AR wayfinding to find the pickup counter makes sense. The app’s already there, you’ve already granted permissions, and you have a specific reason to need navigation right now.

Apple’s Vision Pro and Meta’s Quest 3 both support mixed reality, but expecting customers to walk around a shopping centre wearing a headset remains science fiction. The phone-based approach is probably what we’re stuck with for the next few years.

Accuracy will improve as device cameras get better and the computer vision algorithms mature. The difference between AR tracking on a 2024 phone versus a 2026 model is noticeable—less drift, faster initialization, better handling of challenging environments.

Should Retailers Bother?

For large venues with genuine wayfinding challenges, probably yes. The user experience improvement is measurable, and if implemented well, it reduces load on customer service staff.

For smaller retail, spend your technology budget elsewhere. Better WiFi, functional self-checkout systems, or even just good old-fashioned signage will deliver more value.

The technology works. Whether it’s worth the investment depends entirely on the specific venue and customer base. AR wayfinding isn’t magic—it’s infrastructure, and infrastructure requires ongoing investment to maintain.

Just make sure you go the WebAR route. Nobody’s downloading your app.