On Success and Learning to Fail Well

Do you want to achieve personal and professional success? Are you focused on planning for success? Good. But are you also planning for failure?

You should, because it’s how we fail that sets us up for success, argues David Pablo Cohn. You see, according to Pablo, failure is inherent to the human condition, or, in his words, “Failure is not an option—it’s pre-installed with the standard package.” Which is why planning for failure and learning how to make your failures quick, safe and informative is the secret to success.

It’s a striking message that’s even harder to ignore when offered by a beacon of success like Pablo. With a Ph.D. in Computer Science, the former Google Research Scientist, Instigator at Large and Tech Lead for Google Labs now focuses his attention on philanthropy and providing technical advice for nonprofits. In his spare time he is a flight instructor, travel blogger and, on occasion, tech support guy for the US Antarctic Program.

The intriguing point? He attributes his successes in these and other domains to curiosity, serendipity and primarily the lessons he picked up throughout what he calls “my glorious history of failure.”

We recently had the pleasure of hosting Pablo at Benetech’s offices, where he gave a compelling talk for our team, abundant with examples from Google, aviation and Antarctic expeditions. He shared with us some rules of thumb he’d picked up along the way for failing as well as possible, dusting yourself off and figuring out what to try next. It’s the essence of the Thomas Edison quotation with which Pablo opened his talk: “I’ve not failed. I’ve just found 10,000 ways that didn’t work.”

This was an opportune lesson for the Benetech team. As you may know, we recently launched Benetech Labs, where we explore and test new project ideas so that we can determine which ones have what it takes to deliver social good at scale. It’s where we will conduct our iterative approach to open source development without fear of failure. Put bluntly, with our Benetech Labs’ new project ideas we want to be able to fail often and well, so we can ultimately launch the right tools that help millions of people. By openly sharing that process and what we learn through it, we also hope to assist other social innovators on their paths to set up successful social enterprises.

We believe this could be a useful lesson for you too, with whatever new project idea you’re thinking of or actively working on: be it a new startup, service, product, a book or a charitable organization.

In the spirit of new beginnings and New Year’s resolutions, here, then, are Pablo’s four rules for (good) failure as a way of life:

1. Plan for Failure

Do you think you have a plan? Then start counting the ways it can founder and outline what you will do in each case. Always, always have a Plan B. In fact, having Plans C, D and E would be wise, too!

Think long and hard about all the possible ways your Plan A can fail, and invite both friends and critics to help you enumerate them. For each possible failure, be sure to ask yourself a few questions: Will I be able to detect my plan has failed this way? Is the failure survivable? How can I minimize the expected cost of the failure, and how can I maximize information gained if it does fail in this way?

2. Fail Quickly

Have clear criteria for what counts as failure. Recognize it when it happens and move on. Knowing when something isn’t working and when to quit is crucial: sometimes it’s not just a matter of time and money, but of life and death.

Antarctic explorer Robert Scott was slow to recognize that his plan to reach the South Pole first had failed and he died for his mistake. Ernest Shackleton recognized failure early in his 1914-16 Endurance expedition and his leadership in recovering his crew without the loss of a single life is hailed as one of the greatest survival stories of all time. Shackleton’s expedition, judged by its initial objectives, was a colossal failure, but he quickly turned it into an extraordinary mission of survival and will forever be remembered for his ability to respond to constantly changing circumstances.

Or consider how Chesley “Sully” Sullenberger had only seconds to evaluate his options when US Airways Flight 1549 lost power on takeoff over New York City. His rapid re-evaluation, and decision to land in the Hudson River, was the only option for everyone’s survival.

Failure comes in many flavors. For example, even if a project is a technical success, it will be a market failure if no one wants to use it. Failing quickly means heading off this type of failure before large amounts of effort have been invested. Pablo recommends following Former Google Innovation Czar Alberto Savoia’s approach of pretotyping: finding ways to quickly and inexpensively simulate the project’s core experience with simplified versions of it. “Make sure—as quickly and as cheaply as you can—that you are building the right it before you build it right.”

Embrace failure. Failure itself isn’t your enemy. Denial of failure, on the other hand, is.

3. Learn from Failure

The only complete failure is the one you don’t learn from. After failing, hold a retrospective. Ask what and why— not who—questions. Success follows where failure is seen as a learning opportunity, not a blame game.

Here are some questions to add to your analysis:

–       Was the failure one of those you had anticipated?

–       If not, how well do you understand the system? What other unanticipated failures might be lurking?

–       Are you debugging the right thing? Was it a failure in execution, an unmodeled factor or just bad luck? Bear in mind that bad outcomes aren’t necessarily the result of bad decisions.

4. Iterate

If/when you’ve failed, your Plan B is now Plan A. Do you still have (an acceptable) Plan B? What can go wrong this time? If the failure was not one of the anticipated ones, how are you going to broaden your framework, and what exactly are you going to change so the next failure will be expected? If there is no good Plan B, do you have a pivot?

Pablo also discussed the often-overlooked emotional component to iterating. If you’ve made it this far, you know that failure is… well, hard. Given that 90% of all new start-ups fail, you need to get yourself mentally prepared for the stress and disappointment of failing. Failure causes stress, and prolonged elevation of stress undermines health and well-being. How will you maintain your motivation for the long haul, so you’re ready to throw yourself at the problem again when it’s time?

This is where Claude Steele’s psychological theory of self-affirmation comes in handy. There are many strategies for protecting your health and self-integrity. One proven technique is to keep perspective by reviewing the big picture: Ask yourself, “Why am I doing this?” Write the answer down. Tell your friends and keep a reminder on your desk. Another technique is to look back from the future and ask yourself what you want to be remembered for. Then write down how what you’re doing now helps you get there. Remember the “saying-is-believing” effect, so tell people why you’re pursuing your project. And, of course, a social support network is vital: turn to your family, friends and colleagues for support.

At the close of his talk, Pablo spoke about Google Labs as a model of planning for failure. Google Labs, it turns out, was born in a rather serendipitous manner. It began as a website where engineers could launch “not ready for prime time” projects. But it evolved into a sandbox where Google engineers could conduct rapid experiments and innovate effectively, demonstrating and testing new projects without risking Google’s systems.

With Google Labs, Pablo and his colleagues created an opportunity to fail better—fast, safely and inexpensively—then recover quickly and try again.

Thank you, Pablo, for your inspiring talk. We’re looking forward to implementing the lessons you shared with us in our Benetech Labs work!