I read a fascinating article last week about Google X, the X representing 10; making problems 10 times better in a timeframe of 10 years. This part of the company is challenged to produce “moonshots”. At Google, a moonshot has three criteria. First, it can potentially solve a really big problem for the world. Second, it has to have radical, almost science fiction-like technology. Finally, it has to show progress, giving it a glimmer of hope that it could actually be possible.
A lot has been written about the role of failure in successful innovation. A high tolerance of failure is an essential characteristic of successful breakthrough innovation. This is best exemplified in entrepreneurial start-ups that have a relatively low success rate. There is a tough “survival of the fittest” characteristic. Indeed, savvy venture capitalists look on experience of failure in an entrepreneur’s history as a positive characteristic for the next investment.
When it comes to large organisations, there is much less tolerance of high rates of failure. Shareholders reading quarterly earnings reports generally expect an incremental improvement on the previous quarter and definitely against the quarter in the previous year. This stimulates incremental thinking throughout the whole company. Even if units are established to focus on breakthrough innovation, it’s often difficult to integrate the people and projects into the mainstream organisation.
Expectations are a big issue in this scenario of innovation. If senior leaders at Google expect a high output from Google X they would be sadly disappointed. But they explicitly reward failure.
It’s not really the failure itself that’s important; it’s the willingness to try new things. If you’re scared of failing, you’re less likely to try something new, different and risky. Over time, particularly if management deeds don’t match the words, people become less willing to reach for audacious goals, unless they see an incentive to do so, and are confident that it won’t harm them.
Often in large companies, career development involves a good answer to the question, “what have you launched?”. If it’s your own money and some external investment funds, given in the hope that your risky new opportunity will work, that’s a whole different ball game to working inside a large company with a budget from which a return on investment is a must.
As the wonderfully named Astro Teller, the head of Google X says, “You must reward people for failing. If not, they won't take risks and make breakthroughs. If you don't reward failure, people will hang on to a doomed idea for fear of the consequences. That wastes time and saps an organisation's spirit.”
Teller makes a very good point here about portfolio efficiency. If people are not concerned about retribution for failure, they are more likely to call out a bad project sooner rather than later. That way time, attention and resources can be moved to more attractive options. The efficiency angle also comes through in the “Smartfailing” concept coined by Stefan Lindegaard and Hutch Carpenter. The idea is to find out quickly what is unlikely to work and can therefore be abandoned without wasting too much time and money.
Tolerance of failure is not an acceptance of incompetence. Smart people are still needed, and it’s likely they won’t repeat mistakes or blindly follow routes previously shown not to work.
The lessons for me are:
- Focus on big problems.
- Success is a small number of big initiatives hitting the market.
- Accept a high failure rate.
- Reward people for competent audacity.
Google X clearly doesn’t need to succeed at everything; it just needs one or two. I look forward to finding out what they will be.