XR Adoption Is Moving Faster Than Enterprise IT Budgets Can Keep Up With

Seventy-three percent of mid-size manufacturers in the Asia-Pacific region had at least one active extended reality deployment by Q3 2024, according to a market study released by Delphine Research Group in October of that year. That figure was 31 percent in 2021. The gap between pilot programs and full production rollouts is closing faster than most infrastructure teams anticipated, and that's creating a specific kind of operational stress that doesn't get talked about enough.

Extended reality — the umbrella covering augmented reality, virtual reality, and mixed reality — is no longer a niche experiment. It's a category with measurable enterprise contracts, measurable failure rates, and a body of deployment data that's finally large enough to analyze seriously.

What the Hardware Refresh Cycle Actually Looks Like Right Now

The device side of XR has matured more than the software side. Headset refresh cycles have compressed. Vantix Systems, a Barcelona-based enterprise hardware vendor, shipped its third-generation spatial computing device in February 2025, just 18 months after its second. That cadence is closer to smartphone than to industrial equipment, and procurement teams aren't built for it.

The dominant form factors right now are:

• Standalone headsets running localized compute (used heavily in logistics and field service)


• Tethered mixed reality glasses tied to edge compute nodes on factory floors


• Passthrough AR on ruggedized tablets, which don't get counted in most XR market reports but represent a significant share of active daily users

That last point matters. Tablet-based AR is often invisible in the headline numbers, but Delphine's data suggests it accounts for nearly 40 percent of active XR sessions in manufacturing environments. The flashier headsets get the press coverage.

The Physical-Digital Merge That's Actually Happening in Warehouses

The most concrete deployments right now aren't in consumer entertainment. They're in inventory management, technician training, and remote expert guidance. Korvax Logistics, a Dutch third-party logistics firm, reduced its new-hire onboarding time from 14 days to 6 days after deploying a mixed reality training environment across three distribution centers in 2023. The system overlays picking instructions directly onto physical shelving units using fixed anchor points rather than relying on GPS.

That architecture — fixed spatial anchors rather than real-time mapping — is a design choice that deserves more attention. Real-time environment mapping is computationally expensive and introduces latency. Anchor-based systems trade flexibility for reliability, and in high-throughput warehouse environments, reliability wins.

The physical-digital merge in these settings isn't about immersive experiences. It's about reducing the cognitive load of a repetitive physical task. That's a narrower value proposition than the marketing materials suggest, but it's a real one.

Where the Software Stack Is Still Underbuilt

Hardware is ahead of software. That's an honest summary of the current state.

Spatial content management is fragmented. There's no agreed standard for how XR content gets versioned, localized, and pushed to devices in the field. Meridian XR Platform, which launched in late 2023 and currently serves around 200 enterprise clients, has built a proprietary content pipeline, but it doesn't interoperate cleanly with SAP or ServiceNow without custom middleware. That's a real integration tax.

Analytics is another gap. Most XR platforms can tell you whether a user completed a training module. Few can tell you which physical movements a user made during that module, how long they paused at a specific step, or whether their gaze pattern suggests confusion. That behavioral data exists in principle — the sensors collect it — but the tooling to extract and act on it is immature.

I'd argue the software gap is the single biggest reason enterprise XR deployments stall after the pilot phase. It's not the hardware cost. It's the inability to connect XR activity data to existing operational systems in a way that justifies ongoing budget.

What This Analysis Is Missing

This summary is weighted toward manufacturing and logistics. Healthcare XR — surgical planning, phobia treatment, pain management — has its own deployment patterns and its own set of regulatory constraints that don't map cleanly onto enterprise IT frameworks. Consumer XR, including the spatial computing push from major platform vendors, is a separate market with different economics. And the accessibility picture is largely absent here: most enterprise XR systems assume a level of physical mobility and visual acuity that excludes a meaningful portion of the workforce. Those gaps aren't small, and a fuller analysis would treat them as central rather than peripheral.

The next 24 months will likely clarify which deployment patterns are genuinely scalable. The data from 2023 and 2024 is useful, but it's mostly from early adopters with above-average IT resources. The real test is whether mid-market companies with leaner teams can operate these systems without dedicated XR specialists on staff. That answer isn't in yet.