Plan to Fail Fast by Learning Fast

There’s a lot of talk about innovation in the Federal Government, and with it comes talk of failing fast. The two go together in a useful, integrated way, but I hear little discussion of what either concept means. At least not in terms that managers can use.

A definition of innovation I like is doing something different to add value for a customer. (See Defining Innovation for features of any good definition. And see Why Innovate and Rethinking Government’s Business Model for more on value.) To innovate, you must figure out which different thing to do, to add what value. This also means figuring out what to not do, and that’s where the failing part has meaning.

When applied to innovation, failing fast means quickly shedding things you thought might be solutions in favor of those that will be solutions – or at least those that are more likely to add value for your customer. It’s a process of learning and applying what you learn to build the solution. Think of it as an agile approach to anything, not just software development.

In the commercial world, companies fail fast so competitors don’t add value faster and steal customers. In government, agencies should fail fast to get better ROI, faster, with the investments citizens make in it. If you think of the discovery and learning required to innovate, failing fast as learning fast makes good sense.

So how does one go about failing fast?

  1. Start with a Question. Are you facing a situation, problem, or challenge whose number has come up? Is more of the same no longer acceptable so that now is the time to innovate? If so, give yourself and your team some questions to answer about the situation – rethinking who the customer is, what they value, how you’re delivering it to them, or how you could deliver it (see Start with these Questions).
  2. Plan to Learn. Every successful innovation involves learning. Create space for people to figure out what to do differently that adds value for a customer, and ask them to come back with answers and ideas. If you have one question or goal to learn about, start there but be prepared to reframe your question. If you have a bunch of questions that’s OK, too. Start where you are and let the process shape and focus. Learning will help you see what do to, and what not to do.
  3. Gather Data Quickly. You don’t need years to learn. If your team can devote time to talking to stakeholders, reading reports, analyzing data, and comparing notes, you don’t even need months. With focused effort, you can get preliminary answers to questions in days and weeks. Even if the effort is a large-scale system development or integration, a government team – you don’t necessarily need contractors – can canvass stakeholders quickly. You will gather data continuously, so don’t think of this as one-and-done data collection.
  4. Evaluate Honestly. Regroup with your team and review what you heard. Be relentlessly honest about what the data are telling you so that stakeholder input drives the conversation. It’s OK to question and challenge to illuminate further, but not to reject because you don’t like what you hear. Evaluate honestly to understand deeply, and you’ll elicit information you need to innovate.
  5. Wash. Rinse. Repeat. Plan to iterate with stakeholders so Round 1 insights inform Round 2 conversations, and so on. You can carry these practices through discovery and development….inception, elaboration, construction…whatever process your organization uses. Constantly check what you’re doing differently against what the customer values, and abandon the lesser ideas. Think of it as organizational natural selection. This is how what you learn to not do leads to what you do.

No one wants to fail, and few organizations truly celebrate failure. As more managers and teams practice these activities, organizations will view failure differently. For now, plan to fail fast by learning fast and you’ll add value to your organization and to your customers.

Leave a Comment