What we can learn from the most fascinating spreadsheet in the world

Opinion

Why are only 5% of megaprojects completed on budget, on time or deliver intended benefits? OMG UK’s Phil Rowley condenses Bent Flyvbjerg’s nine pitfalls and their solutions into three takeaways that have implications for marketers.

What’s the most fascinating spreadsheet in the world? Perhaps one that contains the best insight? Or the most revealing truths? 

For me, it’s the dataset gathered by the Danish Economist, Bent Flyvbjerg

For decades now, he has been obsessively collecting data on the delivery of high-profile multi-billion-dollar megaprojects. 

We’re talking about new dams, bridges, airports, rail networks, skyscrapers, or the delivery of mega-events like the Olympics.

Drawing on the wisdom contained in his monster Excel file, Flyvbjerg has written an award-winning book, How Big Things Get Done, a codification of the principles for delivering megaprojects at scale.

I say ‘codification’, but perhaps that’s not quite right, because his rules are reverse-engineered not from the ‘DOs’, but more from the ‘DONTs’ of megaproject misfires.  

Indeed, his database reveals that only around 5% of megaprojects are completed on budget, on time or deliver intended benefits. 

Why do so many miss their mark? 

For many reasons, but I’ve edited and condensed Flyvbjerg’s nine pitfalls, and their solutions, into just three takeaway points that have immediate implications for media and marketing, and their efforts to innovate. 

Although our industry’s hit rate is much higher than 5%, I would argue that we can still learn a great deal about delivery and execution from Flyvbjerg’s “super-spreadsheet” – so let’s look. 

The commitment fallacy

Chief among the many reasons for billion-dollar ball-drops, says Flyvbjerg, is “The Commitment Fallacy”.  

This is the idea that the way to tackle a project is to ‘just get started’ and assume the rest will inevitably fall into place – somehow.  It’s as if symbolically sticking a spade in the ground means 90% of your work is magically done. 

The Commitment Fallacy is built on the notion that publicly announcing an ambitious moonshot heralds ‘crossing the point of no return’. It will generate the necessary motivation and accountability to get it over the line.

But this ‘start now, solve later’ approach is like jumping out of the plane and packing your parachute on the way down, argues Bent. It prioritises the importance of the hoopla around the announcement over the essential work of planning, anticipating and mitigating. 

Berlin’s Brandenburg Airport, for instance, originally set its opening for 2011, but after numerous postponements and revised ‘no this time we really mean it’ launch dates, it finally started taking passengers in 2020 – nine years late. At one point, 120,000 defects were outstanding. 

Is it bad luck or bad planning? Well, Bent’s spreadsheet is clear. Across 95% of megaproject under-deliveries, from broken tunnelling equipment and flooding to staff walkouts and the unearthing of ancient graves, he demonstrates that budget overruns and missed milestones happen not because teams didn’t understand there would be unseen difficulties. 

Instead, it’s that there were no adequate contingency plans for those unseen difficulties.  No “Plan Bs”. No “What-Ifs”. No “War Gaming”. Just over-optimistic haste.

Quoting Abraham Lincoln, Bent reminds us: “Give me six hours to chop down a tree and I will spend the first four sharpening the axe”. 

In short, proper preparation prevents poor performance.

The unique fallacy

Equally dangerous is the assumption that your project is utterly without precedent and that, therefore, any learning from the category or your competitors must be of little use. 

To put it another way, it’s a “reference class error”, where you assume your endeavour demands a new form of classification, rather than typifies an existing one. 

Indeed, Bent shows that so many stadia, subways and skyscrapers have been stymied by the insistence on a ‘unique solution’.  

Using the now-revered, but at the time near-catastrophic, Sydney Opera House as an example, he shows that a feverish pursuit of a one-of-a-kind construction can mean a project gains notoriety for all the wrong reasons: late, expensive, and scrapped. 

But, for Flyvbjerg, the cardinal sin here is not the desire to innovate or do things differently, but the assumption that there are no previous examples from which to draw insight. 

Across the data, his spreadsheet demonstrates that with a wide enough research remit, there will always be precedents, and thus learnings, for project leaders. Even if your endeavours are unique in their combination of elements, it’s likely those elements will have been tested before in other incarnations. 

Analysing those past efforts of others can help you ‘price in’ potential bottlenecks – delays, spiralling costs, labour shortages, tricky planning permissions – even when located in a different project ‘vertical’.

Everything has a reference class. Rail networks can learn about land use from airports. Airports can learn about the price of steel girders from skyscraper construction. Skyscrapers can learn from rail networks about passenger flow. And so on. 

Dismissing the learnings of others because “people who change the game don’t follow the rules” is a risky move.

The complexity fallacy

Lastly, there is the notion that because a project may have earned the prefix ‘mega’, it cannot be broken into discrete elements.  

For Flyvbjerg, some of the most successful megaprojects have thrived because of their ‘modular’ nature. Delivery becomes almost like a production line, pumping out identical, uniform units that are bolted onto the previous one to create scale at speed.   

For example, his data show that megaprojects like solar and wind farms have a storied history of being on time and on budget, precisely because they are essentially one unit, tried and tested, and ‘copied and pasted’ across the landscape. 

Nuclear power stations, in contrast, have numerous, complex, and moving parts. Consequently, they are notorious for astronomical cost overruns and decadal delays. 

Flybjerg, then, encourages us to design our projects from the ‘atoms’ of small testable technologies, asking us to ‘Find The Lego’ that corresponds to our vision.  

His go-to example is The Empire State Building, which was constructed out of a simple, repeatable floor shape, jacked up into the sky, before the next floor was copied and pasted in underneath.  

Despite its shape being a synthesis of form, function, and construction, it remains iconic and perhaps the most famous building in the world.  Notably, however, the Empire State Building not only went on to be the tallest building ever constructed, but more importantly, was delivered early and under budget. 

Innovation does not necessarily mean complexity.

Applications for media innovation 

Whilst it would perhaps be a stretch to describe any innovation initiative in media, marketing and advertising as a megaproject, I still think there are some fascinating takeaways from Professor Flyvbjerg’s work. 

First, Plan. Elevate the value of scenario forecasting, or as we refer to it at OMG Futures, Prospective Analysis or ‘Pre-Mortem’. 

It’s vital to actively anticipate potential issues with efforts to innovate, whether due to budget cuts or changed timelines. The question we pose to brands is usually, “Imagine it’s 12 months from now and you have not innovated in media and marketing. Tell me the reasons why.” The act of marketers picturing their reality with potential future blockages helps them envision solutions. Remember: commitment to innovation is nothing without contingency. 

Second, Learn. Consider brand case studies as the “York Notes” for our industry. Learn as much as you can about media innovation inside your category, and out. Study the reference class and know what benchmarks to aim for.

Furthermore, never dismiss a media innovation because another brand has already executed it. You are permitted to adopt, provided you also continually adapt.

Learn from others’ successes, but more importantly, their mistakes. As the saying goes: “The early bird catches the worm, but the second mouse gets the cheese”. So, develop a “second mouse playbook” to improve upon others’ benchmarks.

Third, simplify. In response to Bent’s plea to ‘modularise’ innovation, think about how you normalise small test-and-learn projects that cumulatively build experience and insight over time. 

Media innovations do not always have to be the unique, disruptive, big blow-out element of your campaign. Instead, innovation can be a chain of tried and tested elements, delivering success via ‘compound interest’ -all while flexing the innovation muscle.

Incidentally, we’ve written extensively about how to put these principles into action in our e-book, How To Innovate.

The big takeaway from How Big Things Get Done is that we should treat innovation like infrastructure. 

From AI ads to airports, and from skyscrapers (banners) to skyscrapers (real ones), the principles and pitfalls for innovators remain the same. If you want to deliver on your ambition, big or small, then plan, learn and simplify.   

Phil Rowley is head of futures at Omnicom Media Group UK and author of Hit the Switch: The Future of Sustainable Business. He writes for The Media Leader about the future of media.