On Data: The Proposition 

One ongoing challenge we’ve long faced is that often technology outpaces progress. Many enterprise digital transformation projects begin with good intentions, only to lose focus and spiral out of control. Despite process mapping and updating standards to come up with a solution that meets specific business needs, along the way the technology and capabilities become the focus of the project.  Often, selecting a product without first having a plan can have an adverse effect on its entire purpose.  In this blog post, we’ll discuss the factors that hinder success when implementing technology. We’ll help you recognize some of the key causes of failure and give you the steps to take appropriate action.

The Wish List Problem

It is necessary for asset-intensive organizations to identify and implement solutions that effectively enable business processes.  Not doing so can adversely affect their competitiveness and lead to higher costs and inefficiencies.  The market, in this age of artificial intelligence, bombards buyers with messages about capabilities and innovations, positioned as solutions that solve a range of business needs.  Buyers will assemble a wish list citing capabilities such as the following, which are copied from real-life requirements documents:

• AI, machine learning and advanced analytics to process large datasets, identify patterns and drive predictive and prescriptive maintenance strategies with automated insights and trend analysis.
• Data analytics systems that use advanced analytics tools, such as statistical analysis, machine learning algorithms and data visualization tools, to analyze the data and identify patterns and trends that may indicate potential issues.
• The system uses advanced analytics and machine learning algorithms to analyze collected data and identify patterns and anomalies that indicate potential equipment failures.

My first thoughts when I see these lists are:

1. “Do you fully understand what you're asking for?” and  
2. If these capabilities were already in place, how would you recognize or measure their impact?"

Lists of desired requirements, like the examples above, sometimes ask for dependencies that exceed the buyer’s technical boundary.  Responding vendors may not be able to answer the questions as the requirements are not self-explanatory.  This has always been a challenge facing buyers and vendors, but without widespread artificial intelligence, identifying the true enabling capability is impossible.

The Reality Behind "Prescriptive Maintenance"

Prescriptive Maintenance is usually on these lists of requirements and I’m choosing to discuss it because I believe it is a conceptual placeholder.  But even so, we need to consider what something like prescriptive maintenance would entail.  First, consider the massive data needed and the effort required to coordinate equipment telemetry, operational context, cost visibility, supply chain integration, and other records. Operationalizing the data is one thing, but choosing the appropriate data model to use it requires a level of dedicated expertise that most asset-intensive organizations don’t have.

This scenario brings to mind an episode of Young Sheldon, a television prequel series that tells the fictional story of Dr. Sheldon Cooper, theoretical physicist from the hit show, “Big Bang Theory”, as he grows up as a boy in East Texas. Young Sheldon has an eccentric personality, high intelligence and struggles with social cues.  One day, he builds a small nuclear reactor to power his home and neighborhood. His motivation for this is that his father had scolded him for wasting electricity by staring into the open refrigerator. While Sheldon’s goal was simply to stand in front of the fridge as long as he wanted, he didn’t fully consider all the stakeholders involved —like his parents, the Atomic Energy Commission and the FBI. Even if he had succeeded, his plan wouldn’t have easily fit into the right operating or business context.

When Innovation Actually Works

It’s true: sometimes innovative technology exposes and solves a problem that we didn't know we had. Consider the iPhone: could anyone have predicted that in the near future it would be possible to summon a taxi at the click of a button on your phone?  

This was one of those pivotal moments where a foundational innovation triggered an era of continuous progress over the next 12 to 15 years. The era of Industrial Internet of Things, Industry 4.0 and edge computing followed soon after, with the market driven by the idea that every device on earth would have an IP address and could be interrogated.

The Transformation Failure Rate

Research conducted by Accenture in 2022 found that seven out of 10 enterprise transformation efforts fail to fully meet business leaders' expectations.  There are numerous reasons for this.  Let’s explore this further.

The Fitness Industry Analogy

The implications shape the expectations of people in asset management, including maintenance and reliability. When looking to how they manage asset management performance, they often look to a market that promises far more than what is realistically achievable.

For example, this can be seen in the fitness industry, where social media influencers with flawless skin and perfect muscle tone sell an unattainable ideal.  This is a fitting analogy for the "digital transformation" era of the last 15 years; much of it isn’t real and even if you do buy into it, you’ll probably spend money with little to nothing to show for it.

The Gap Between Vision and Reality

The gap between the sophisticated "prescriptive" maintenance and current capabilities is significant. Most organizations are still grappling with implementing effective predictive maintenance, let alone achieving true optimization-based prescriptive approaches.

Terms like "prescriptive maintenance" or "digital twin" are often a conceptual placeholder for advanced capabilities that the industry aspires to achieve, rather than describing widely available, proven technologies. It's more accurate to think of these as emerging categories with a lot of development ahead of them, as opposed to established, standardized practices.

AI that can reliably optimize complex maintenance decisions across multiple variables, costs, risks and operational constraints is still mostly theoretical or exists only in very limited, controlled applications.

The Path Forward

Complexity is inherent in asset management, but it doesn't have to be overwhelming. The first step is recognizing the difference between achievable progress and overhyped promises. Let's instead talk about addressing the gap between the aspirational "vision" sold by the market and the practical, achievable steps organizations can take to improve asset management performance.

Rather than chasing the latest technological buzzwords, organizations should focus on understanding their actual needs, building solid foundational processes, and implementing technology that serves those processes—not the other way around.

This is the first in a blog series exploring practical, data-driven asset management strategies that cut through the hype to deliver real results.