Before I start, I want to thank Sam Gregor, Building Structures Digital Lead at Mott MacDonald, for agreeing enthusiastically to collaborate on this short piece. It is always a pleasure to share thoughts with you!
Failure is the rule, not the exception
In the book How Big Things Get Done, the authors Bent Flyvbjerg and Dan Gardner highlight a sobering truth about large-scale projects across industries: failure is the rule, not the exception. They call this the Iron Law of Mega Projects, which states that most large projects exceed their time and budget and underdeliver on the promised benefits.
This isn’t just anecdotal—data backs it up. In their analysis of 258 transportation projects, only 0.5% were delivered on time, on budget, and with the intended results. The average project experienced budget overruns of 28%, with rail projects faring the worst at 45%. Even worse, when they expanded their research to include major IT initiatives, some projects saw costs exceed expectations by 447%.
From infrastructure failures like the Berlin Brandenburg Airport in Germany to IT disasters like the initial rollout of healthcare.gov in the United States, these patterns persist across countries, sectors, and project types.
While failure is unavoidable in complex projects, its role depends on timing and context. In early design stages, failure often serves as a tool for discovery and iteration, but in later phases, particularly in high-stakes construction, “right first time” becomes essential.
The findings from How Big Things Get Done are especially poignant for the architecture, engineering, and construction (AEC) industry. The sector operates under immense pressure to deliver projects that balance safety, predictability, and efficiency while adhering to tight budgets and timelines. This creates a paradox: while failure is a pervasive reality, culturally and organizationally, failure is not an option.
Failure in this space can be a disaster, bringing serious consequences like delays, rising costs, and a hit to reputation. In high-stakes environments—where lives, reputations, and financial liabilities are on the line—playing it safe is not just understandable; it’s often essential. However, while this cautious culture may be justified in some instances, it frequently extends unnecessarily into other domains, such as digital solution development, where iterative approaches and experimentation are critical for success. The challenge lies in knowing when to play it safe and when to embrace calculated risks. Timing and context are everything.
Early in a project, particularly during design, testing, or prototyping phases, taking calculated risks and surfacing potential failure modes is critical to uncovering weaknesses and refining solutions.
If a 10% increase in design costs leads to a 10% reduction in construction costs—a phase that typically dwarfs design expenses—the value is undeniable.
Wolf Mangelsdorf – Professor for Structural Design, Die Angewandte
However, as projects progress to execution—whether constructing a bridge or deploying life-critical systems—the focus shifts to precision and reliability. The problem arises when this “right-first-time” mentality unnecessarily bleeds into digital solution development, where iterative approaches and a willingness to experiment are vital for success.
This fear of failure often leads to a culture of excessive caution, focusing more on avoiding risk than embracing new ideas. Conversations within organisations rarely dig into the root causes of ongoing budget and schedule problems. Instead, public discussions usually celebrate a project’s successes—like how it will improve accessibility or boost the economy—while brushing aside the challenges and mistakes faced along the way.
A missed opportunity to learn
The AEC industry needs to take opportunities to learn and improve. While other sectors use postmortem reviews well, they’re uncommon in AEC. This leads to repeating the same mistakes, which keeps things inefficient and causes delays.
In aviation, the focus is not on who is at fault but on why a failure occurred. This blameless postmortem concept ensures that every incident—no matter how minor—is meticulously analysed to improve future outcomes.
Some people in the AEC field might think the term “postmortem” sounds too harsh and want to use more positive wording. But changing the language doesn’t tackle the main issue: projects continue to go over budget and miss deadlines, yet we still celebrate when they’re finally done, acting like everything went perfectly.
Exclusively praising results fosters overconfidence in poor practices, reinforcing the status quo instead of driving improvement. Social scientists have found that when accountability is tied only to outcomes—success or failure—teams are more likely to persist with deficient strategies, even when evidence suggests they are not working.
Ironically, an industry focused on minimising risk consistently aligns with the Iron Law of MegaProjects—delays, overruns, and underperformance. With 90% of projects failing to meet their original objectives and a reluctance to critically examine failures, the tendency is to move on to the next project, perpetuating the cycle.
Fail Fast
The fear of failure affects how we approach projects and permeates internal development as well. This fear often causes citizen developers to delay deployments, waiting until they’re sure everything is perfect. But this cautious mindset can be counterproductive. If you want to ship successfully, you must do the exact opposite: deploy as early and as often as possible. Delaying deployment delays potential failure. But if failure is inevitable, you want this to happen early. Alternatively, this fear could mean that promising projects are never even started, missing out on successes or, at worst, learning opportunities.
Teams that start with reliable bi-weekly releases often progress naturally to weekly, then daily, and eventually multiple releases per day. Each update becomes small, independent, and atomic. This practice, called Continuous Delivery, benefits teams and customers, enabling faster feedback and improvements. Despite its proven advantages, many organisations believe that caring about quality requires slower, quarterly releases. The book Accelerate offers compelling evidence that Continuous Delivery leads to better quality and faster results.
Tackling the scariest changes first is also key. Early in a project, you have the most evident context and understanding of the system. Delaying complex updates shifts responsibility to others who may need more familiarity, increasing risks and slowing progress, aka failing slowly. Remember, failure isn’t a necessary evil—it’s a necessary step toward doing something new.
SpaceX exemplifies this philosophy. Their approach to rocket design prioritises rapid iteration over rigid, linear development. Instead of spending years perfecting every detail before building, SpaceX begins with a clear goal and quickly moves to concept designs, prototypes, and testing. Their mantra is simple: build and test early, find failures, and adapt. This iterative process allows them to identify problems sooner, make adjustments, and innovate faster.
We must recognise that failure is not a negative thing but a sign that your team is exploring uncharted territory. Mistakes are signals of experimentation and innovation. The goal isn’t to avoid failure entirely but to fail intelligently—learning and refining the process to minimise future errors. Growth happens through this trial-and-adjustment process.
Pixar also offers a shining example of how organisations can embrace failure. Ed Catmull, one of its co-founders, cultivated a culture where mistakes were seen as part of the creative journey. His philosophy was simple:
Making mistakes is the privilege of the active.
In contrast, bureaucracy thrives when people fear decisions, avoid responsibility, and stall progress.
Conclusion
In the AEC industry, failure is often seen as a weakness, something to hide rather than embrace. However, this perspective limits growth and innovation. To truly evolve, we must change our mindset, shifting from avoiding failure to failing better, earlier and faster. When handled intelligently, failure becomes a powerful tool for discovery, learning, and improvement.
Failing well doesn’t mean being reckless. It means taking smart, calculated risks and documenting both successes and mistakes.
We celebrate our wins publicly and document our failures and lessons openly.
This transparency creates a foundation for shared accountability and continuous learning, where every misstep becomes a building block for progress.
This requires a significant cultural shift in the AEC industry. Failure, when it happens early (i.e. before it threatens life, the environment or economic viability), should not be viewed as a disaster but as an opportunity for reflection, iteration, and growth. Teams must feel empowered to ask: What went wrong? What can we do better next time? Normalising these conversations encourages introspection, innovation, and better project outcomes.
The Path Forward
To embrace this new mindset, organisations must focus on three essential steps:
- Fail Better: Celebrate failures as part of the journey, reframing them as opportunities to learn and grow. Documenting mistakes openly fosters transparency and accountability, paving the way for meaningful improvements.
- Fail Early: Encourage teams to tackle challenges and test assumptions early when the risks are smaller, and the lessons learned can shape future iterations. Early mistakes cost less time and resources and provide critical insights that accelerate progress.
- Create Safe Environments: Build environments where teams can test ideas without fear or anxiety, allowing digital and product teams to experiment freely. These spaces should remove barriers and limitations, enabling developers to adopt new technologies and methodologies without hesitation.
Failure isn’t the end—it’s the beginning of discovery.
A huge thank you to Ben for his invaluable support.