How to learn to find viable hypotheses and prioritize them

[sakibkhan22197]

Formulate a hypothesis: based on the collected data, propose a specific change. For example, “Adding auto-fill of the address to the order form after selecting on the interactive map will increase the order conversion from 5% to 10%.”
Example of a funnel for an online store and conversions in it
Such a funnel for an online store can be built in Carrot quest
Don't know where to start building a funnel on your website or the current model doesn't suit you?
Contact our experts for a free consultation. They will help you with this and tell you how to improve the conversion of the funnel on your website.

Sign up for a consultation
Testing hypotheses takes time and resources, so it is important to make only promising assumptions before testing. Here are the signs of a working hypothesis:

The result is measurable. For example, it is difficult to evaluate the list of afghanistan cell phone number effectiveness of a lead magnet compared to a discount on a course. But for an initial assessment, you can compare the conversion to contact from pop-ups with a discount promo code and a lead magnet.
The goal is achievable given the budget and time. If previous campaigns have shown that discount pop-ups are more successful, it is worth testing them, calculating the advertising budget and setting aside time for design and analytics.
The wording includes a test element and a target metric. For example, "a pop-up on the home page will increase clicks on the button from 1000 to 2000."
The hypothesis influences the main metric. Based on the results of the hypothesis testing, you can understand how the target indicator has changed.
It is important to formulate hypotheses for a specific market segment. Because each industry will have its own customer journey, competitors, tools, sales channels, and metrics. Analyze the customer journey using CJM (Customer Journey Map) , and based on the results, formulate hypotheses for the following tests.

Growth teams start working on hypotheses with research. They collect data from the marketing funnel, assessing at what stages the conversion rate decreases. For example, site users come to the blog to read articles, but do not leave a request in the form after the text. Then the guys generate hypotheses that could increase this conversion. Such brainstorms are held once a week. Then there is a meeting with the manager, at which they defend the hypotheses themselves, key metrics for them and what will be considered a good result. And then the teams go into a sprint for implementation.

Hypothesis Prioritization Methods
ICE
The method works best if you have a list of hypotheses and want to quickly assess which ones might provide the most benefit given limited resources.

How to use: