The steps of conducting growth experiments are generally divided into: generating experimental ideas, sorting experimental ideas, designing experiments, developing experiments, analyzing experimental results, and producing growth reports. Below we introduce them separately.
1. Generate experimental ideas
1. Clarify the objectives of the experiment
Our growth experiments must be based on what we found.
Is a button conversion rate too low? Or is a feature that has a low retention rate? Find the problem, we can know the direction of optimization. For example, we found that the conversion rate of the registration function is only 30%, then we can formulate our experimental goal as: increase the conversion rate of the registration function from 30% to 50%. Below we all take the case of low conversion rate of registration function as an example.
2. Find the cause of the problem
In order to achieve the experimental goal, we have to find the cause of the problem.
We can find the cause of the problem in both qualitative and quantitative ways.
Qualitatively, we can understand why the registration function conversion rate is so low through user research, user usability research, etc.
On the quantitative side, we can gain insights through data. For example, through the funnel model, it is possible to see where users have the highest churn rate; through user grouping, whether to know which users have high registration conversion rates, which types of users have low registration conversion rates, and so on.
3. Find solutions based on the cause
Knowing the cause of the problem makes it easier to find a solution.
We can find solutions qualitatively, quantitatively, and best practice.
Qualitative and quantitative will not be introduced, which is similar to "find the cause of the problem".
Talk about best practices here. Best practices are those methods that are recognized as likely to increase conversion rates. Such as B=MAT model, Lift model.
In the B=MAT model, to improve the user's conversion rate, we need to give users sufficient motivation, reduce user resistance and improve ability, and country email list give trigger reminders in appropriate situations ( trigger).
The Lift model is mainly divided into Value Proposition, Relevance, Clarity, Anxiety, Distraction, Urgency.
value proposition. With a clear and powerful marketing slogan, users can accurately perceive what benefits they can get.
Correlation. Landing pages and conversion pages meet user expectations and are closely linked to your value proposition.
clarity. The experience process is smooth, and users clearly know what to do next.
Anxiety. Do subtraction, don't give users too many choices; don't do things that don't meet users' expectations and habits.
distraction. Reduce visual contamination, information noise.
urgency. Create a sense of hunger to enable users to make decisions quickly, and be good at using the endowment effect and loss aversion.
Here we use the above method, assuming that the solution for "low conversion rate of registration function" is as follows:
Shorten the unnecessary registration process on the page, for example, users are not required to fill in personal information such as email addresses during registration. Just enter your phone number and verification code to log in and register.
Highlight the sign up button.