Skill Level Intermediate
- If you ever feel like the product or feature you're working on is a leap of faith and needs a prayer, then it's time to use a hypothesis. A hypothesis is an idea or explanation for something that is based on limited facts or evidence. It needs more investigation to be generally accepted. Hypotheses are ways for teams to setup a structured and controlled investigation. It's a way to learn while making progress towards the goal and while reducing risk. Also, hypotheses are great as an overall technique to keep an eye on the outcome and metrics you're looking to move.
Teams use hypotheses in some common situations to help innovate. They use them when driving innovation to ensure that only build enough at a time, so they can learn from each piece and reduce the risk of rework. Many hypotheses lead to additional experiments or sub-experiments. Teams also use hypotheses to remove ambiguity in both business and technology assumptions or leaps of faith. Let's look at how a typical hypothesis is written up.
We start with a "we believe that" statement, and specify the feature or item we're building. Then we specify for whom we're building it for. Next, we have a "will achieve" statement, with the desired outcome or measurement that we want to go with and get to with the feature being built. And last, we have a "we will know we're successful when" statement. And this one has a metric or a threshold value we need to get to.
So, for example, a team is working on implementing an image recognition and analysis technology to identify who is entering the building. Currently, they use a badge and turnstile system, and would like to automate further to remove the badges and turnstiles, reducing the risk of shared badges and compromising security. A hypothesis this team might develop to work on before building the entire solution, might be something like, "We believe that using facial image "recognition to identify your employees "when entering the building, will reduce "fraudulent entrances and increase "building security.
"We know we're successful when the accuracy "of this technology alerts security "of fraudulent entrances at above a 99% accuracy, "and accurately identifies our employees "99% of the time." A team would also then define the controlled experiments that will help them learn if their hypothesis is on track. Hypothesis driven work is an important piece of today's innovation processes. And a technique that will be critical to a team's success.
Leap into using hypotheses and keep your team on track.