How to formulate a hypothesis and test whether it will work

[sakibkhan22197]

For example, an online school launches a hypothesis: "Game mechanics with a pop-up on the main page will help collect 10 times more leads than other mechanics on the site." Marketing tests this hypothesis on a group of students and tracks changes in sales. If the experiment shows a positive result, the offer is scaled up.

We didn't come up with the hypothesis from the example above. It was actually launched in the online school Moscow Digital School and 108 payments were collected in 3 months .

Want results like Moscow Digital School?
Start testing hypotheses with Carrot quest tools. Use ready-made list of afghanistan cell phone number templates of chatbots, pop-ups and email newsletters to launch the first mechanics and collect applications in just 7 days of the trial.

Test Carrot quest
Hypotheses arise from ideas, but unlike ideas, they are concrete and measurable. Hypotheses are always linked to business goals and are based on specific metrics. For example:

"If you reduce the number of steps in the form, the conversion to registration will increase" - that's the idea.
“Reducing the registration form by one step will increase conversion from 5% to 10%” is a hypothesis.
To turn an idea into a hypothesis, you need to go through four steps:

Define the goal: specify what result we are aiming for. For example, the goal “to collect more leads on the site” sets the direction, but is not a hypothesis in itself. It lacks a specific area that we influence and a metric by which we will understand that the result is good.
Select a metric within the goal: study the customer journey and determine which metric to influence. Example of a strategy: we look at the sales funnel, namely the conversion rates when a deal moves from one stage to another, and notice the areas where conversion drops the most. These areas of the funnel are called "narrow" and are the ones where improvements are suggested. Everything is simple here - working on a "problem" area is more likely to give a better result than on one where conversion drops almost imperceptibly. The decision on which stage of the funnel to work with can be made based on the team's strategic goals for the year.
For example, in an online school, a sharp drop in conversion can happen after a demonstration or a trial lesson - users are not sure that they want to take this particular course or doubt the price. Working with the audience and warming it up before buying in this area will most likely bring results in contacts and applications left.
Assess the potential for improving the metric: Based on data and analytics, suggest how the metric can be improved. For example, our conversion rate from a completed demo to an application is 15%. The research we study may show that in our industry this conversion is twice as high, therefore there is potential for growth of 15%.
Find ways to influence the metric: explore what changes can improve the situation. Let's say a customer survey revealed a problem with filling out a course questionnaire. It has a lot of questions that no one fills out completely. Improving this area could be the key to increasing conversion to purchase.