A few years ago, conversations about XR training centered on hardware. We talked about whether enterprise content could run on consumer headsets, how to manage those headsets, and how to roll out content across a distributed fleet.

The technology matured rapidly, and content developers kept pace. Developers quickly stretched beyond simple simulations, incorporating principles of instructional design for content that engaged and stuck with learners.

As MDM solutions like ArborXR made device and content management easy, early adopters eagerly launched pilot programs. In a 2024 survey by Training Magazine, nearly a quarter of large enterprises reported training people using virtual reality. ArborXR is currently partnering with more than 3,000 companies, including more than 60 in the Fortune 500, on XR programs.

XR training has proven its scalability. It’s proven its viability. Now, it must prove its value. In this new age, it won’t be enough for content to be beautiful, scalable, and engaging. Content must gather and relay data that proves the program’s value as a learning tool.

Talented developers continue to reach new heights in content quality. But by and large, the data element has been missing. Managers at the top of early adopter organizations are asking about their return on investment (ROI). In many cases, XR program directors are coming to the uncomfortable realization that they don’t know the answer.

Pressed by the needs of their customers, some developers have stepped up to fill the gap. They’re not only integrating data collection into their content, they’re also creating and maintaining the infrastructure that makes that data useful.

The result is that developers whose time and resources are best spent creating and iterating on training content are instead tied up on questions of data storage, security, and visualizations customized to each client.

The challenge facing the XR training industry today is cracking open the black box: getting quantitative data to prove ROI without building custom layers of high-maintenance data infrastructure. ArborXR is rising to meet the challenge.

ArborXR Insights makes it easy to capture learner data, present it in clear reports and dashboards, and integrate it with existing Learning Management and Business Intelligence systems.

Without Data, XR Training Content Is Built on Guesswork

Some XR teams bristle at the assertion that they’re not collecting data. To be sure, many collect anecdotes and user surveys to see how content is performing. Updates and refinements are based on this user feedback.

But when the finance team is allocating next year’s budget, qualitative data and sentiments don’t cut it. Stakeholders need to see quantitative data to assess the impact of the deployment and compare it objectively to traditional training methods.

Some XR teams get creative.They find things they can measure and guess at whether those metrics are representative of program success.

One of the better examples I’ve seen was a team that tried to prove ROI by comparing the cost of shipping a headset to a learner against the cost of flying the learner to a training facility and putting them up while they took in-person training.

That provides some rough math on expenses, but it gives no measure on whether the learner actually understood and retained the material.

I’ve also encountered teams that collect metrics manually. Trainers watch the learner through a screencast of what’s in the headset and make notes on their performance. This method doesn’t scale, and it doesn’t make a strong case for why the organization should keep investing in VR.

One reason it’s so hard to measure XR training is that it’s not clear what to track. There’s no established playbook of XR training metrics. The measurements that matter when training in a hard skill like operating machinery might not apply to soft skills like negotiation.

Another challenge is authentication flow. Identity management within the headset is a consistent struggle. When data is fragmented across apps, event tracking and reporting for individual learners at scale becomes a monumental task.

The Trouble With Current Data Solutions

When developers build data collection into their content, it’s usually only half a solution. The data exists, but only within the vendor dashboard. It’s cut off from business and learning systems, preventing the organization from getting a complete picture of its training program.

XR teams resort to manual tracking, CSV exports, and custom pipelines in an effort to bridge the gap. Sometimes, a single individual is responsible for aggregating and reporting data across apps.

These manual workarounds are prone to breaking, create bottlenecks, and limit the organizational impact of the XR program. Frustrated developers are put in the role of data babysitters, creating and maintaining custom pipelines instead of innovating new content.

Besides being fragmented and high maintenance, custom pipelines tend to be vulnerable. Improvised data integrations rarely meet security standards. And since they exist exclusively within the vendor dashboard, if an organization changes vendors, or the vendor shuts down, previous training data is lost forever.

When developing ArborXR Insights, we wanted to futureproof data integrations. We developed a tool that reports on data across content platforms, simplifies authentication out of the box, and integrates with more than 500 LMS and BI dashboards for holistic insights.

How Missing Data Affects The Ability to Prove ROI

ROI is a hybrid metric. It’s not just a question of whether it’s cheaper to ship a headset than to fly a learner cross-country. It’s a question of aggregating all the program data: costs, benefits, effectiveness when compared against other modalities, and integrating it into the big picture.

Without reliable, accurate data on both individual and program outcomes, it’s almost impossible to justify continued investment in XR training over traditional learning methods. The inevitable result will be programs getting shut down or treated as a novelty, stuck in permanent pilot mode with no path to scale.

What Good Looks Like: A Blueprint for XR Training Data

With ArborXR Insights, XR training content becomes measurable and scalable.

• Simple authentication - launch directly from your LMS or sign in with a headset PIN.
• Track what happens in the headset - completions, scores, and custom events across all apps.
• Powerful dashboards - easy-to-read reports that show learner progress and outcomes.
• Seamless integrations - export data automatically to 500+ LMS and BI tools.‍
• Fast setup - simple and powerful SDK. Start tracking basic events like completions with a few lines of code.

Building the Playbook

With Insights, we can start building that standard playbook of XR training metrics. Tablestakes metrics are when a user starts and finishes a training; these data points alone will give teams deeper understanding of their program.

It takes less than five minutes to add Insight’s SDK to content’s code and begin tracking these basic metrics. The resulting data can be set up to flow directly into a BI or LMS dashboard.

From there, organizations can track whatever metrics they need to accurately assess the impact of their content, such as telemetry, custom events, and interactions.

By deciding what to measure at the beginning, and building content to track it, we start to build a shared language of XR training data. The more consistent we are in talking and thinking about these metrics, the easier it will be for the industry to define “good” in broad categories of XR training.

What to Track When Measuring XR Training

The metrics that matter depend on who you are. Content providers care about different analytics than educators, who care about different things than enterprise L&D teams. Understanding the goals of the people measuring the program is the first step in deciding what to track.

Content Provider Goals: Reduce Friction and Improve Product

Content providers are concerned with the user experience inside the headset. Tracking user paths and dropoffs will help identify areas of confusion that need to be smoothed out.

In both reality and virtual reality, people don’t typically move directly from Point A to Point B. They take detours and encounter obstacles along the way.

At a minimum, content providers need to track where users make right choices and where they make wrong choices. Additional data points, like where they are in the space and how much time lapses between events, can add context to the picture.

Telemetry is another useful tool to content developers, alerting them to technical issues within the content or device. Tracking performance in real time is invaluable to debugging to improve the user experience.

Educator Goals: Personalize Learning and Improve Retention

Educators want students to engage with their learning and get the most out of the experience. Contextual data is important.

A student who logs a lot of wrong answers but takes their time getting there is probably struggling with the material. A student who logs a lot of wrong answers really fast is just trying to finish the module as quickly as possible, clicking around until they happen upon the right answer.

Tracking interaction frequency, the number of failed attempts, and the time on task gives teachers valuable information about a student’s learning style.

Telemetry and assessment data can be used to adapt content so learners get extra time to work on difficult areas and progress quickly across areas of aptitude. We can expect that level of personalization to improve retention and reduce frustration.

Enterprise L&D: Prove ROI and Scale

In business, learning and development teams need metrics as soon as possible so once the pilot ends, they can prove ROI and get buy-in to scale.

“You need an immediate metric after the pilot,” said the CEO of a leading immersive training platform. “You can’t say, ‘Well, in six months, we’re going to find out what happened with this one.’”

Decision makers need to be able to compare XR training against existing training modalities. Identify how traditional programs were being measured and incorporate the same metrics. L&D teams likely need to track completion rates, scores, attempts, and time to complete.

Decision makers determining whether to continue or scale a training program need the data put into the context of business impact. To make the data easy for executives to see and understand, stream it directly into LMS and BI dashboards.

The XR Industry Needs a Mindset Shift From Delivery to Data

The decision of how to measure training can’t be tacked on at the end of the development cycle. Data instrumentation needs to be intentional. If you’re building training without instrumentation in mind, you are flying blind.

Training development can take a page from product development: understand the job to be done and what success looks like from Day One. Then build the training to achieve that goal. This mindset lets us take inspiration from product analytics tools like Pendo for how to track and interpret user behavior.

Data will guide content providers in developing content that reflects the way people learn. Training development is never one-and-done. It’s a feedback loop. By incorporating data collection from the outset, developers won’t have to rely on surveys and anecdotes from small pools of users. They can log decision points, track replays, identify patterns, and compare completion against comprehension to determine if training really worked.

We built ArborXR Insights to enable this mindset shift. We want you to confidently iterate on content, smooth friction points, increase engagement, and make your training genuinely valuable to learners and organizations.

Getting Started: The Path to Value

Before developers begin data instrumentation, they need to understand what their customers want. What metrics do they want to measure? Do they require an LMS or BI integration? Which pieces of existing content do they want to retrofit with data tracking?

With that understanding, implementing ArborXR Insights is simple. The system was created to deliver quick wins with built-in SOC 2 Type II and ISO 27001 compliance and secure, flexible deployment.

For content providers, a quick win might be collecting insights for customer reports. For educators, the priority might be instrumenting time on task tracking. For enterprise L&D teams, it could be streaming completion data to BI.

In each case, start with EventAssessmentStart() and EventAssessmentComplete(). With those two lines of code, you can start tracking assessments. Within 15 minutes, your XR training data can be integrated with your LMS dashboard. Add SDKs to track additional events as you identify them.

Where Does XR Training Go From Here?

The XR training industry is growing up, moving from a period of experimentation into a more sophisticated era of application. Experimentation will certainly continue to play a role, particularly in the area of artificial intelligence; I think adaptive learning based on AI feedback is on the near horizon. But by and large, we need to shift our attention from whether the content we create is innovative to whether the experience we provide is valuable.

Let’s take user authentication as an example. Imagine being an elementary school teacher with 20 children in headsets, all trying to type in their passwords on clunky virtual keyboards. The more time it takes for users to authenticate their profile and get to the training, the less time they have to engage with the content itself. ArborXR Insights enables the technical backend of the content to fade into the background from the user perspective, even as it collects powerful information.

Until now, developers have been understandably focused on building innovative learning modules. But for the industry to grow and adoption to increase, we need to consider the entire session—from how a user logs in, to how the data is stored to their profile, to how their profile data is aggregated into program analytics. Expanded investment in XR programming will be a much easier sell if each step flows smoothly and easily into the next.

This new era also demands flexible applications that can adapt to future advances in technology. AR and MR will have increasingly important roles in training as technology evolves. When building a VR application, it will be important to consider how this application would perform in an AR or MR environment.

The most important thing to remember is that no advancement will be possible if developers cannot prove the effectiveness of their content. Without reliable data, there is no way to convince investors of ROI.

Until now, the XR training industry was small enough that content providers could get away with not tracking analytics. Those days are gone. Data is the key to unlocking scale. Over the next five years, developers who cannot prove their outcomes with data will be left behind.

ArborXR Insights makes XR training measurable, credible, and scalable. Data is the bridge between innovation and impact, and the organizations that cross it first will define the next decade of learning.