Innovation in Software

No matter. Try again. Fail again. Fail better.
– Samuel Beckett

Whenever I am asked by a client why Agile Development is a good idea, why TDD or BDD is the way to build a product, I can sum up the argument pretty succinctly: it reduces the cost of change to as close to zero as possible.

Recently I have been doing some reading around “Lean Development”. I have found Eric Ries’ Lessons Learned blog an invaluable resource for finding jumping off points into resources and ideas that are still very new and fresh to me. This morning whilst leafing through my new copy of The Principles of Product Development Flow, which I bought after seeing it recommended by Eric a few weeks ago, I realised I could succinctly reduce the arguments for lean development down to another mantra:

Lean Development reduces the cost of failure to as close to zero as possible

It doesn’t sound like much of an insight, does it. In fact, everything you read about this area is about testing Potentially Fatal Assumptions (PFAs), and market-validating feature sets: get your product out of the door, don’t invest too much in case it fails, if it succeeds then well done for lucking out. And of course, lean development is not just about failing at things – it’s about reducing the cost and time to delivery whilst also focusing on the only factor that really counts in product development: profitability.

However, when you let the idea about cheap failure seep in, you realise it challenges deep-rooted beliefs held in our society.

From the nursery school chart of “gold stars” all the way through our continuous grading of achievement in our first 20 or so years of life, most of us have learned through trial and error that failure is bad. Failure is (literally), for losers. Failure is not something you want associated with you, your ideas, your relationships or your life. Failure, we believe, is bad.

I think we’ve taken a wrong turn somewhere. Providing that the cost of failure is low – in terms of capital, resource, emotional investment, etc. – failure is useful. Failing quickly at something that was always going to fail, allows us to direct our attention to something new.

Our lives on this planet are short. We can only fit in so many projects, businesses and relationships in our average 70+ years on this planet. Why on earth should we invest a sizeable percentage of that precious fleeting existence on something that is doomed to failure from the outset? Why do we fight on investing more emotional and financial resource into something that will never ultimately work?

So, test for failure early, quickly, and cheaply. Sure, you can try and work out whether you need to riff through something because you’re in The Dip and that might work for you. What I’m saying though is that when you fail at something quickly and cheaply, you have opened up your life to a new possibility, a new project or idea that isn’t flawed and will be successful.

I’m not saying you should aim for failure. I’m saying that if something is fatally flawed, find out quickly, and rejoice. If it’s a success, well, brilliant: that was the point all along.

In project terms, I’ve failed plenty of times. My only regret is that I didn’t fail more cheaply. I invested so much emotionally, physically and financially that my resources were depleted for the next project, which in turn may have caused failure where there wasn’t a fatal flaw. From now on, failing early and failing often will be a sign for me that I’m getting better at understanding fatal assumptions. One day I will run into an idea that thanks to my cumulative experience of flaws will have none that effect its profitability, and it will hopefully succeed in ways I can not yet imagine.

When you realise that something is failing, resist the urge you’ve had all your life to try and “turn it around”. Find the PFA that is no longer just a potential, learn from it, and get on to the next project. Quickly.

Footnote: I remembered whilst writing this post, a friend who got divorced less than a year after getting married. He and his wife found that their attitudes to starting a family had significantly diverged, and so realised (and I quote), “it’s best to just call it a day early, otherwise we could lose years of our lives, wasting them on a relationship that was never going to last long-term” (emphasis mine). That took some guts, however their lives will be the better for admitting the fatal assumption that they both wanted the same things going forward, and they will reap the benefits of not sinking so much of their lives into something fatally flawed. If they can be so mature and pragmatic about something so emotionally raw, I’m sure you can be about your next idea for an iPhone app nobody cares about.