Titans of innovation

AS A technical feat, ATLAS takes some beating. It is the world’s biggest microscope, used by physicists at CERN, a large laboratory near Geneva, to probe the fundamental building blocks of matter. Its barrel-shaped body, 45 metres long, 25 metres tall and weighing as much as the Eiffel tower, was assembled in a cavern 100 metres beneath the Swiss countryside from 10m parts, nearly twice as many as in a jumbo jet. It generates more data each day than Twitter does.

It is also a remarkable organisational achievement. The components were designed by hundreds of scientists and engineers from dozens of institutions. They were subsequently sourced from 400-odd suppliers on four continents, at a cost of $435m. At any one time the experiment involves more than 3,000 researchers from 175 institutes in 38 countries.

Does a multinational science project like ATLAS have much in common with a multinational business? Some management types assume it does. Occasionally they offer to advise the boffins how to run it better. Alas, their advice is “too abstract and difficult to understand”, scoffs one physicist. ATLAS, for one, seems to be doing just fine without it. Last year—in case you missed the headlines—it nabbed the Higgs boson, known as the “God particle”. It could be that Big Science has more to teach big business than vice versa. A recent Strategic Management Society shindig at CERN and IMD business school in Lausanne drew a sell-out crowd of academics, business folk and consultants.

“Big Science” projects differ from companies in important ways. They are publicly financed and do not seek profits. They are also one-off affairs, with no need to maintain supply chains or manage long-term relationships with customers. Yet, like companies, they must innovate furiously, make the most of limited resources and beat rivals to breakthroughs.

Although no two big experiments are exactly alike, all share important traits, says Philipp Türtscher of the Vienna University of Economics and Business Administration. Their aims are often clear-cut—find the Higgs, sequence the genome, potter around Mars—but the means of attaining them are anything but. To shorten the odds of success, individual design decisions are put off for as long as possible. This, says Markus Nordberg, who co-ordinates ATLAS’s finances, lets the project “absorb uncertainty”. Companies, by contrast, typically try to reduce it by picking one solution that is known to work and sticking with it.

In a Big Science project, teams with rival proposals spar publicly, forcing all the boffins to articulate their assumptions, justify their choices and learn enough about their rivals’ ideas to criticise them at length. One engineer who had left industry to join CERN was bemused, recalls Mr Türtscher, when he told his physicist colleagues that a cooling system could not be made any smaller and they had the temerity to ask, “Why?” (On closer inspection it turned out that the device could be shrunk by a fifth.) The natural inclination of scientists to challenge authority is given free rein in a big collaboration because individuals are ultimately accountable to (and paid by) their parent institutions. This explains why experiments have “spokesmen”, not managers.

The sparring takes a while, but the lifetimes of Big Science projects are measured in decades. Besides, a good scientific scrap fosters the exchange of ideas and ensures that advocates of losing proposals understand the winning ones, which they are expected to work on. Battles allow boffins to let off steam, so grudges seldom fester. The bonhomie of Big Science even rubs off on commercial partners: when one firm working on a NASA space telescope ran out of beryllium for components it was making, another offered some of its excess stock.

One way for companies to emulate Big Science is to employ more scientists. Simon Williams, co-founder of QuantumBlack, a London-based data consultancy, says his visits to CERN (to seek technical help with number-crunching) prompted him to value PhDs over MBAs. They can be a handful, Mr Williams concedes, but they also require less hand-holding. Give them an interesting problem and they will get cracking, he adds with enthusiasm.

Hiring eggheads rather than dunderheads is generally wise, though it can backfire: just ask the banks that employed “quants” by the dozen to create financial instruments that no one understood. As a rule, firms can attract megabrains only if the problems they want to solve are interesting. Basic science meets this criterion; people become scientists to extend the frontiers of knowledge. For many, it is an obsession. The only banter Schumpeter heard on a visit to the CERN canteen concerned physics. The television screens were all tuned to live feeds from experiments—on a Saturday morning. Few firms inspire such commitment; and no mission statement or desk massage will change that.

Suits follow lab-coats

Nevertheless, many innovative firms appear, consciously or not, to have recreated aspects of the Big Science model. Google and 3M encourage workers to spend some office hours tinkering with pet projects. Infosys, an Indian IT company, gets people from different parts of the company to vet each other’s ideas. Valve, a big American maker of computer games, eschews hierarchy in favour of majority decision-making, including to select new staff. AT&T even had a pure-maths department. The snag is that only the best companies can enjoy the luxury of picking the best brains and letting them roam.

Managers with liberal-arts degrees are not wholly useless. The pointy-haired boss in the “Dilbert” cartoons, who once tried to improve a product design by scribbling out a timing circuit and writing in “Moby Dick by Charles Dickens”, is, fortunately, a caricature. But the suits have much to learn from the lab-coats.

^{Economist.com/blogs/schumpeter}