Superconnect, By Richard Koch and Greg Lockwood

This primer on network theory fails to convince

Network science sure is popular. Linguists, biologists and computer scientists churn out statistical models and theories on the interconnectedness of their respective fields. At its best, it can produce beautiful graphical analyses of how things work. At its worst, it can produce utter banalities. So open-ended is the idea of network science that it was only a matter of time before it was seized upon by a self-helpish author with an eye for a buck. And indeed it has come to pass, with Richard Koch and Greg Lockwood's new book, Superconnect: The Power of Networks and the Strength of Weak Links.

Richard Koch made his name in the 1990s writing books about the 80/20 rule – that 20 per cent of inputs create 80 per cent of results – based on the work of the 20th-century economist Vilfredo Pareto. Pareto did the number-crunching; Koch turned it into a bestseller.

Superconnect is aiming for a similar market. The book's central proposition is this: we naturally operate within a close network, be it home or work, where our bonds are strong and continually reinforced. However, when we want to produce something new, or move job, it is the weaker connections – those people we rarely contact – that end up being most productive.

Beginning by trying to convince the reader that there was something miraculous about the authors' own meeting – they both, through a series of coincidences, ended up sitting on the board of Betfair – hardly inspires confidence in the scientific basis of this book.

Superconnect then proceeds to access a wide range of data to bolster its argument. The authors turn to Stanley Milgram's 1960s "small world" experiment, which asked people to forward a letter through a series of known friends to reach a target. From this somewhat primitive piece of research, we can see how far our network cobweb spreads.

Then we must understand the hub. Take Denis Diderot. The editor of the first encyclopedia ran himself into the ground writing it, with the help of only 100 co-contributors. By comparison, Jimmy Wales, the founder of Wikipedia, breezed to success in a mere 12 months, with 150,000 voluntary aides. Ergo, Wales is at the centre of an expansive network helped along by the powers of the internet. Like Google and other big players, Wikipedia has become a hub which continues to strengthen links with its users as it grows. And Diderot was a failure?

And now the crunch. We are asked to believe that some of the strongest connections actually come from exploiting the weaker links between various hubs. The evidence here begins to thin, as Koch and Lockwood come up with a series of spurious personal examples of how a distant connection often provides a leap, such as a new job. (The book doesn't mention that someone looking to fill a post might well pick someone they barely know because their faults aren't so obvious.)

If we are yet to be convinced, the theme is padded out with numerous cultural references – to the Jewish faith, James Dean and Google – alongside hat-tips to serious network theorists such as Albert-Laszlo Barabasi, all contrived to fit the argument.

This book is an exhortation to turn away from your normal hub, which provides security in its familiarity, and to exploit crossover points in your network through those long-neglected contacts. Think of the people on Facebook – the ones you barely contact. If Koch and Lockwood have it right, hammer them with a few messages over the new few days, inveigle yourself into another network, and what will likely follow is a new job, business partnership or relationship.

But take these 300 pages of advice as one might read a horoscope. Superconnect relies extensively on anecdote and, where real scientific research is used, it is so uncritically done so that I would remind you of another far more succinct truism that your parents probably handed down to you years ago: "Don't go into business with friends or family." It works just as well as a rule of thumb. Better the devils you don't know.