Learning Wit

Fail fast.  It’s one of the Agile buzz phrases that gets thrown around a lot in software product organizations these days.  Particularly, organizations trying to embrace the Lean/Agile approach to production.  The term ‘fail fast’ is grounded in the Lean concept of continuous learning.  Lean theory contends that learning is not a singular event, but rather a continuous process of trial and error.  The Lean approach advocates that the smaller the ‘set’ of learning and the faster it takes place, the better.  Thus, fail fast should really be ‘learn fast’ or ‘learn something small fast,’ but that’s not nearly as catchy.  This is all grounded in the heavily researched area of human learning.  Humans learn by trial and error.  Lean simply says so should the organizations.

Learning is Feedback

A fundamental element of learning is feedback.  You must know you’re doing something wrong in the first place.  When building a product, the ultimate source of feedback is customers.  In building software, the faster and more feedback you can get, the more learning that takes place.  To do this, you need the mechanics and infrastructure in place to facilitate getting feedback.  The ability to push product updates quickly and measure the impact of those updates is critical in software design.  Consumer-oriented web offerings have become masters at this, with the ability to push hundreds of small updates daily.  Enterprise-grade web applications may not be that fast yet, but they are catching up.  Being able to be able to measure and monitor product performance and user interactions in near real time is also a great source of feedback.  Having data to measure is essential to learning if you hit your target or not.  The mechanics to test new ideas in ‘live’ environments without hurting customer productivity if you break a feature is extremely helpful for learning.  Often called ‘beta testing,’ having an isolated and safe environment to test product features and stability gives you plenty of room for making mistakes, and learning from them.  These are straightforward and widely used concepts in many software product organizations today.  However, far too many companies still lack these mechanics, and without them, feedback and learning typically do not happen, and it certainly does not happen fast.

Culture of Learning

Having the mechanics and infrastructure is critical, but without a culture oriented around learning, even the best measures won’t help you.  If learning is about ‘trial and error,’ then you must allow for the ‘error’ to happen.  That means being wrong, and in today’s business environment, being wrong is not exactly something celebrated.  You have to shift the conversation away from ‘right vs. wrong’ and towards; what have we learned and what can we do to get better?  It’s why doing things in small sets, quickly, is so important. Better to course correct quickly on small mistakes, than to wait six months to figure out you’ve totally missed the target.  This requires allowing work to be organized into small sets, moving quickly, and being comfortable with early work that is unfinished or unpolished.  The standard should be “good enough to get feedback on” and not “done and perfect.”  Leaders should encourage and celebrate learning, regardless of whether the outcome supports a particular point of view.  Creativity, innovation, and imagination thrive when the threat of punishment from being wrong is removed and individuals are provided the freedom to learn.

The Foo of Learning

At WitFoo we believe strongly in metrics and measuring everything we do.  We have a passion for learning, to get better, and to continuously improve.  We are not worried about always being right. In fact, we expect that in most instances we start off wrong.  As we measure and collect customer feedback, we learn and draw confidence knowing we will end up in the right place.  We have a saying at WitFoo when starting something new. “Make it fast, make it messy.”  We worry about getting it right only when we can measure feedback from our customers.  How well is it working?  Time will tell.  But, given the amazing work being done by this team of scrappy professionals, I like our odds.

Share This