Why Digital Twins Fail (And How To Avoid It)

The digital twin market is experiencing explosive growth. The industry is expected to grow from $24.48B in 2025 to nearly $260B by 2032. And there are a host of reasons why.

Digital twins provide operational visibility. They enhance collaboration. They make it easier to make high velocity, high quality decisions. And they unlock opportunities for predictive maintenance and other strategic initiatives. McKinsey suggests the upside of a digital twin implementation is material, with improvements in capital and operational efficiency of 20-30%.

And yet. A major roadblock to achieving the promised land of digital twin adoption is the sheer complexity of it all. Success requires a delicate interplay between strategy, technology, data integrity, and organizational readiness.

In this article we outline the primary reasons for failure, and provide guidance on how to avoid a similar fate.

The Three Primary Reasons Digital Twin Initiatives Fail

For the purposes of this article, we’ll define failure as an inability to generate a positive ROI within a reasonable timeframe.

This matters, because many firms will suggest simply getting a digital twin off the ground is success. But our clients (usually owners) care about outcomes. It’s not enough to get it live. It has to get results.

Third party research validates what we’ve seen in the field for years. There are a set of challenges that can be grouped into distinct but interconnected themes. Being aware of them and developing strategies for proactively mitigating them are critical.

While not trivial, technology is actually the simplest part of the equation. The bigger challenge is usually in aligning the technology with the people and processes it is meant to serve.

Reason 1: No Clear Goals or KPIs

It’s hard to get an ROI if you haven’t defined what success looks like. The most common barrier we see is an inability (or unwillingness) to define specific, measurable business goals. Without them, the digital twin becomes a fancy but ultimately underused asset.

If you don’t know what you’re using the twin for, it’s borderline impossible to stumble your way toward meaningful ROI improvements. What gets measured gets managed.

A good example comes from a recent engagement with Disney, which was developing a new facility. Their facility leaders knew from experience that poor record documentation would create downstream headaches during operations. So they defined clear, measurable goals years before completion. They asked:

1. How can we get a 3D model of the facility at the end of the project that allows us to locate assets, see behind walls and ceilings, and trace systems?
2. How can we ensure that model is accurate?
3. How will we maintain and update those models in operations, either internally or through a vendor?
4. How can we make those models accessible and integrated with other building systems?

Those might sound simple, but they did the job. In fact, simple KPIs are often idea, because everyone understands them. Clarity, alignment, and the ability to measure progress are what we’re after here.

How to address it:

Start with a business case. Define the specific problems you aim to solve and the KPIs you’ll use to measure progress. For some owners, that means focusing first on documentation accuracy and accessibility; for others, it might mean speeding up maintenance workflows or improving energy performance. Make sure the goals are explicit, clear, and measurable.

It’s important to recognize that your starting point will shape your goals. For existing facilities you’ll likely focus on comparative data, or how your facilities perform relative to each other. For new facilities you’ll likely focus on things like accuracy and accessibility, making sure you have usable documentation from day one. Make sure you select goals that make sense for where you’re at.

Reason 2: A Brittle Data Foundation

The most sophisticated digital twin is only as reliable as the data that feeds it. And often that foundation is sorely lacking. A 2023 literature review identified static building data as a primary barrier to digital twin adoption.

Many owners struggle with getting their data foundation in place because there’s no internal champion or “data governor” responsible for alignment. It’s critical to have someone to connect everyone across systems, workflows, and business units.

Organizations also often struggle to plan for the data proliferation a digital twin makes available. A digital twin doesn’t generate most of this data. But it does bring it all together. Data comes from BIM models, point clouds, laser scans, reality capture, and on-site photo documentation. And once in operation, even more data is brought in from IoT sensors, building management systems (BMS), and ticketing or maintenance platforms. It’s a LOT. It’s important to have a foundation in place that can scale as your digital twin matures.

Lastly, interoperability can be an issue. Digital twins are effectively “systems of systems”, needing to pull data in from a dozen or more places to unlock their full potential. Depending on the systems you’re using and the capabilities of your team, stitching all these together can be a challenge.

How to address it:

Adopt a maturity model. You don’t have to do everything at once. You can move up the levels of digital twin maturity in a phased approach, getting your data house in order as you go.

We generally advocate for a process that looks something like this:

• Foundational Twin. An inventory of building components and baseline information (often drawn from BIM models, asset data, and reality capture) that tells you what is there.
• Descriptive Twin. A shared, data-rich visualization of the physical environment (often layering in data from Building Management Systems (BMS), Computerized Maintenance Management Systems (CMMS), and discrete IoT sensors) that tells you what is happening.
• Integrated Twin. A connected, intelligent system that shares data between operating platforms bi-directionally (often incorporating AI or machine learning to interpret inputs and identify causal relationships) telling you why it’s happening.
• Predictive Twin. Reaching the ability to forecast tasks through real-time feedback loops, telling you “what will happen.”
• Prescriptive Twin. Over time, data insights are fully integrated into a learning environment where the twin tells you “what should be done”, and in some cases even automates it.

To be candid, most organizations are slow to get to levels 4 or 5. That’s okay. You can generate sufficient ROI at earlier phases, and don’t need to wait to do so.

Reason 3: Change Management

Lastly, organizations often underestimate the change management side of the equation. Now, this is completely normal. Change is inherently hard, because it runs counter to how our brains are wired. As Bent Flyvbjerg notes in How Big Things Get Done, people consistently underestimate the time, effort, and friction involved in adopting new ways of working.

(Note: this problem is only compounded by a well-documented skills shortage. Honeywell points out  digital twins are actually a solution to this problem. But it will take work to get there.)

Again, an internal champion is invaluable here. They make sure the project maintains momentum, all key players get what they need, and maintain data documentation integrity throughout. They also help drive adoption in day-to-day workflows by getting feedback feedback from end users, finding areas of friction, and making sure the twin becomes a regular part of how their teams work.

Note that change management isn’t just an issue up front, but remains an issue on an ongoing basis. Buildings are constantly evolving, and one of the first questions owners ask when planning a digital twin is, “How do we keep it up to date?” Too often, there’s no plan or resources for maintaining data accuracy over time.

How to Address It:

Have a comprehensive change management plan that covers both people and data. Invest in upskilling your workforce. Create cross-functional teams including IT, operations, and facilities. Make data governance part of your operating rhythm. And make sure your plan doesn’t just address launch, but factors in ongoing stewardship.

Conclusion

Digital twins can revolutionize how your organization operates and manages its portfolio of buildings. But getting there requires proactively addressing complex realities of implementation.

If you’d like help designing a strategy for making Digital Twins a reality in your organization, we’d love to talk.