3.4. Hypothesis — Data Driven Learning

From the moment we open our eyes in the morning, we make decisions. Should I lie for another 5 minutes or get up right now? Should I drink coffee or tea? Should I watch some news on TV or read a newspaper? A countless number of decisions are made every single day by people of all ages, races and incomes, in all parts of the world. But how to make a right decision when there’s too much put on stake?

How NOT to Make a Business-Related Decision

We need to make a decision. A decision related to our business. Maybe we need to add a new feature or a new button, or maybe we would like to add a new plan to our pricing page. Should we do that? Or maybe that’s not such a good idea and we should leave everything as it is? How to decide?!

So what should we do? How should we make a decision?

How to Make a Business-Related Decision

Decision should be made based on data. There are two types of data: soft and hard. Soft data can be obtained from customer development, it’s based on qualitative interpretations, assumptions, observations, and opinions. Hard data is always factual, sometimes technology-generated, when information is presented in the form of quantifiable results (figures, numbers, graphs, tables, etc.). We need all data. The more data there is, the less is the chance of making a mistake. Accordingly, the first thing you need to do to make a decision is collect and analyse data.

After you’ve collected all the facts, opinions and other kinds of data, you need to make a reasonable assumption of a specific result, i.e. you need to generate a hypothesis. Here you can find all information on the hypotheses and their kinds. An important point to keep in mind is that a hypothesis should always be measurable (compare: “If we change the colour of the button from red to green, we will increase conversion from 12% to 20%” vs “If we do something, something good will happen”). If it’s measurable, it is manageable.

What Next?

You’ve collected data, generated a hypothesis. What next? Firstly, if you have a lot of hypotheses, then you need to prioritise them, because it will be too problematic to test them all at the same time. Accordingly, choose the best ones, the ones that you think will be important for your business. After we’ve gathered some information and generated a hypothesis, it’s time for experiments. Experiment with the best hypotheses! Hypotheses make you think, experiments provide you with knowledge. Having knowledge and data, you are given an opportunity to look into the future and learn from it.

Only… Who Can Work With Hypotheses?

Depending on the structure of your company, there are different options who can generate hypotheses and work towards growth. If you’re a young startup or a small- to medium-size company, you can choose the person who has a wide area of knowledge, is responsible, creative and thorough. That person should know how to change, where to change, and how to organise changes. If your business is big and you have what is needed to hire a growth hacker or even a whole team — don’t doubt it. By investing into a growth expert once, you can increase your income and awareness of your whole business for a long time!

To Sum Everything Up…

Regardless of the structure of your company, anything can be a source of hypothesis. Cases, examples, advice, research, reports, ideas — everything. You need to be able to see opportunity everywhere. Moreover, you shouldn’t think that only top levels of management can provide good and valuable ideas — the lowest persons in the company should also be asked and heard. By the way, the sign in hotels about towels on the hook/on the floor was invented by a simple cleaning lady who earned $3/day. Ideas come from everybody in your team, you just need to listen!