The Cost of Fragmented Monitoring: Why Projects Still Fail to Turn Data into Decisions

Here is a number that infrastructure professionals understand intuitively but rarely see quantified: 91.5% of megaprojects exceed budget, schedule, or both. Not a few. Not half. Nearly all. This figure comes from Bent Flyvbjerg’s analysis of over 16,000 projects—one of the most comprehensive datasets on project performance. 

While these overruns are often attributed to planning, scope, or execution challenges, one underlying factor consistently emerges: how monitoring data is structured, shared, and interpreted.

McKinsey reports that large projects typically take 20% longer than planned and can run up to 80% over budget. Meanwhile, a joint Autodesk and FMI study (2021) estimates that poor data practices contributed to approximately $1.8 trillion in global losses in a single year. 

This is not a failure of effort or expertise. It is a systemic issue. The way information flows—or fails to—creates a structural disadvantage. At its core, a significant share of project overruns can be traced back to fragmentation.

What Fragmentation Actually Looks Like?

At the end of a long day, a project manager reviews a 100-page compliance report: graphs from inclinometers, piezometers, strain gauges, and settlement markers. Each dataset comes from a different subcontractor and is delivered on different timelines in different formats.

The task becomes simple: check thresholds, confirm compliance, and move on.

This is not negligence—it is a rational response to a system not designed for decision-making.

This is monitoring as a retrospective exercise. It shows what happened but rarely what to do next. It does not connect signals:

The industry excels at collecting data. We can measure millimetric strain across kilometers, monitor ground movement via satellites, and stream sensor data in near-real time.

What has not kept pace is the ability to organize, correlate, and interpret that data for decision-making.

The result is decision latency—the gap between when data signals a problem and when engineering action is taken. When that gap stretches from minutes to days, costs compound silently.

In projects using the observational method, where design evolves based on real-time data, this delay is not just inefficient—it is a design risk.

Read more: The Shift from Fragmented to Integrated Monitoring

What Does Each Stakeholder Lose?

Asset Owners 
Without integrated data, prioritization becomes guesswork. Every asset appears equally urgent—or equally stable—making it difficult to allocate resources effectively.

Consultants 
Disconnected data removes critical feedback loops. For example, inclinometer data on a metro project revealed that conservative grouting was contributing to deviations. Without integrated insights, the default response remains overdesign.

Contractors 
Time is lost navigating fragmented information instead of acting on it. The data exists—but the pathway from signal to decision is unnecessarily complex.

What Works: Prioritize, Instrument, and Cross-Verify?

The solution is not more sensors. It is a structured approach:

1. Prioritize at Scale

Before deploying ground instrumentation, satellite-based screening establishes system-wide awareness.

On a major metro extension in the Middle East, over 300 structures across 20 stations were screened via satellite. The corridor was divided into 14 zones and classified by risk.

This allowed the client to focus ground monitoring only where necessary—improving both efficiency and clarity.

In another case, a mountainous road network required additional environmental validation due to vegetation interference. Of seven initial zones, two were reclassified as stable—ensuring sensors were deployed only where truly needed.

Read more: Why Is Fragmented Data Killing Your Productivity (and How to Unify It)?

2. Instrument Where Risk Concentrates 

Once risks are identified, instrumentation becomes targeted. 

Instead of deploying sensors based on uniform specifications, placement is guided by data. 

Twenty well-placed sensors can deliver more actionable insight than 200 deployed by default. 

3. Cross-Verify Through Convergence

The true value of monitoring lies in data convergence.

It is not about having multiple instruments—it is about correlating them:

When signals converge, confidence increases. 
When they diverge, early warnings emerge.

On a deep metro excavation (20–24 meters depth), this approach identified a critical issue:

Four datasets, one coherent story.

The solution—adjusted dewatering, additional wells, and controlled excavation—was driven by system-level understanding, not isolated readings.

This principle applies across project types.

On a cable-stayed rail bridge in the Himalayas:

A key insight transformed the monitoring strategy:

Every structure has a natural frequency signature.

Using six accelerometers and FFT analysis, engineers tracked structural behavior over time. Matching frequency responses across sensor pairs confirmed structural symmetry, with accuracy rates exceeding 95%

Six sensors, combined with intelligent analysis, delivered deeper insight than hundreds of isolated measurements

The value was not in volume—it was in convergence.

The Opportunity Ahead

Most projects today have monitoring systems. 
Few have integrated decision workflows.

The industry has spent decades building the instrumentation layer—and it works.

The next step is integration.

This approach must be embedded from the start: 
Prioritize from space → Instrument where risk concentrates → Converge to uncover the full picture.

The cost of fragmentation is not just delayed, or budget overruns.

It is the loss of confidence in decision-making—and the missed insights that never reach the engineers who need them most.

By Jose Antonio Berrio (VP - International Services)

FAQs

1. Why do most infrastructure projects exceed budget or schedule?

A major reason is not just planning or execution but fragmented monitoring data. When information is disconnected, decisions are delayed, leading to compounding risks and costs.

2. What is data fragmentation in infrastructure monitoring?

Data fragmentation occurs when monitoring data comes from multiple sources, formats, and timelines without integration. This makes it difficult to connect signals and derive actionable insights.

3. How does fragmented data impact decision-making?

It creates decision latency—the delay between identifying a risk and taking action. This delay increases the likelihood of cost overruns and design inefficiencies.

4. What is decision latency in infrastructure projects?

Decision latency is the time gap between when monitoring data indicates a problem and when corrective action is taken. Reducing this gap is critical for project performance.

5. Why is collecting more data not always the solution?

More data without structure increases complexity. The real need is to organize, correlate, and interpret existing data effectively for decision-making.

6. How can infrastructure projects reduce monitoring inefficiencies?

By following a structured approach, prioritize high-risk areas, deploy targeted instrumentation, and integrate multiple data sources for unified analysis.

7. What does “data convergence” mean in monitoring systems?

Data convergence refers to correlating multiple datasets—such as displacement, pressure, and load—to validate patterns and improve confidence in engineering decisions.

8. How does satellite monitoring support infrastructure projects?

Technologies like InSAR provide large-scale displacement data, helping identify high-risk zones before deploying ground-based sensors.

9. What role does monitoring play in the observational method?

In the observational method, design decisions depend on real-time data. Fragmented or delayed data can introduce serious design and safety risks.

10. What is the future of infrastructure monitoring systems?

The future lies in integrated systems where data is connected, analyzed collectively, and translated into clear actions—reducing uncertainty and improving project outcomes.