What 22,000 Projects Reveal About Success and Failure

Why It Matters

Understanding that true project success hinges on value creation, not just on‑time‑on‑budget delivery, forces organizations to invest in rigorous planning, ultimately improving ROI and reducing waste.

Key Takeaways

• Planning investment of 20‑30% dramatically improves project outcomes.
• Only 0.5% of projects meet budget, schedule, and benefits simultaneously.
• Success can occur despite overruns; value creation outweighs cost overruns.
• Agile without thorough front‑end planning risks accelerating failure.
• Historical data from 22,000 projects highlights importance of realistic forecasts.

Summary

The Shift Code podcast hosts Alexander Budzier, an Oxford professor, discussing his research on 22,000 projects to redefine what constitutes project success. Rather than relying solely on the traditional Iron Triangle of cost, schedule and scope, Budzier argues that a project should be judged by the net value it creates relative to its investment.

The data reveal stark realities: roughly half of all projects stay within budget, but only about eight percent meet both budget and schedule targets. When benefits are added to the equation, a mere 0.5 % of projects achieve all three criteria. Nonetheless, about half of the initiatives generate enough value to be considered successful, even if planning was unrealistic.

Illustrative cases underscore the paradox. The Sydney Opera House exceeded cost and time limits yet transformed the city’s economy, while Lidl’s €500 million IT platform delivered on schedule and won awards but was abandoned for lacking business fit. Budzier also cites the Madrid metro and a nuclear plant where extensive front‑end work proved essential.

The findings compel leaders to allocate a larger share of resources—often 20‑30 % of total cost—to thorough planning and realistic forecasting. Agile methods that skip this stage risk technical debt and failure, whereas disciplined upfront analysis can boost ROI and reduce the odds of costly overruns.

Original Description

In this episode of The Shift Code Podcast, host Pierre Le Manh is joined by Alexander Budzier, Fellow in Management Practice at the University of Oxford's Saïd Business School, to share insights on what project success actually means and why traditional metrics fail to capture it.

What You’ll Learn:

- What truly separates successful projects from slow-motion failures.

- Why the best leaders focus on building systems rather than giving firebrand speeches.

- How projects can succeed spectacularly on one dimension while failing on another.

- The uniqueness trap as the single biggest red flag for cost blowouts.

- Why investing 20–30% of the total cost in planning dramatically improves outcomes.

- How AI is already boosting individual productivity by 40%, but only 5% at the corporate level.

Alexander Budzier is a Fellow in Management Practice at the University of Oxford's Saïd Business School, where he researches and teaches how to set major projects up for success. His dataset of more than 22,000 capital investment projects across 126 countries is the empirical foundation for two books published by Wiley in 2025: How to Measure Anything in Project Management, co-authored with Douglas Hubbard and Andreas Leed, an Amazon number-one bestseller in software and IT project management; and Intelligent Change: The Science Behind Digital Transformations, which draws on Oxford research to establish what actually drives success in digital programs.

His research has appeared in the Harvard Business Review, where his article on uniqueness bias, co-authored with Bent Flyvbjerg, was selected for HBR's 10 Must Reads 2026, and in leading journals across management, energy, and transport. He is CEO of Oxford Global Projects, a specialist advisory firm, advising governments and major program teams across transport, infrastructure, IT, and energy. He completed his doctorate at Oxford in 2015, having previously worked at McKinsey's Business Technology Office in Düsseldorf and Chicago.

Highlights and YouTube Chapters:

[00:00] Introduction

[01:15] The Intellectual History of Project Value

[06:04] The Sydney Opera House Paradox

[10:18] What 22,000 Projects Reveal

[14:04] Think Slow, Act Fast

[18:49] Where Complexity Hides

[25:09] Three Pitfalls of Forecasting

[29:30] The Uniqueness Trap

[32:27] The Power of the Pause

[39:43] What Leaders Must Unlearn

[44:13] AI in Project Management