How to set your team up for failure

FAIL by Fuzzy Gerdes via Flickr

Failing is vital. If your only plan is to succeed, you’ll never learn, and thus never have the crucial breakthroughs your product needs. Many enterprises only value success, so we fall into a pit of setting low goals to ensure success and hiding failures. In his excellent Mind the Product talk, Dave Martin goes more into why accepting failure is vital to a culture of innovation – https://www.mindtheproduct.com/2017/05/building-a-culture-of-innovation-through-product-management-dave-martin/

The greater danger for most of us lies not in setting our aim too high and falling short; but in setting our aim too low, and achieving our mark.
-Michelangelo

So, how do we set our teams up to fail, when they’ve been in cultures of success for years? For me, explicitly calling out expected failures and a little bit of story structure goes a long way. Here’s what I do:

  • Tag or name user stories that may fail as experiments. This helps the team open their mind to being risky and think about the minimal amount of work to test an experiment.
  • Create two additional stories for every experiment. One story is the additional support we’ll do if the experiment is a success, like hardening the code, adding more tests, and checking performance. The other story is what we’ll do if the experiment fails, such as reverting tests and code. Ultimately you’ll delete one of these stories and execute the other.
  • Ideally, these two follow-up stories should have the same estimate. This shows that the appropriate amount of effort is being put into the experiment and you’re not over-investing in the success or failure route. An ideal experiment has a 50/50 chance of success to optimize learnings, so we want both routes to have the same cost. If you need to track the total cost for an experiment, take the more expensive path of success or failure (not the cost of all three stories).
  • Once you’ve done your first story and measured your results, delete one of the stories, and play the other one, and share the learning from the experiment’s success or failure with others.

Hopefully you find this framework for tracking experiments helpful too. It helps the Team get in the mindset for failure, and creates the plan to fail so they’re ready to aim high and learn together.