Designing with data.

Data is crucial to optimizing the design process of improving customer value and engagement.

The following is a 10-step process we use to enhance the product design using data.

1. Problem statement.

2. Hypothesis.

3. Audience.

4. Primary success metric. 

5. Target lift.

6. Counter metric.

7. Baseline conversion rate.

8. Analyzing the experiments.

9. Experiment results should be documented.

10. Designing with data.

 

1. Problem statement.

An explanation of the internal business or user problem you are trying to solve.

Today’s case study focuses on the fitness industry and was developed by Amplitude, and the main business problem that needs to be improved:  how can we increase the number of classes booked?

This case study explores the following questions:

How can we increase class booking conversion?

How do we convert non-power users to power users?

2. Hypothesis.

An assumption of what actions could be taken to solve or alleviate the problem statement and why.

The experiment was designed to test this hypothesis :

By getting users to invite friends to workout together we can increase the number of classes they complete in a week.

 

3. Audience. 

A group of users that will be targeted for the experiment. This audience will typically be split evenly into “control” and “variant” groups.

 

4. Primary success metric. 

The main metric you hope to move by running this experiment. Should ideally drive both customer and business success.

The primary success metric for the non-powers users is the average classes per week.

 

5. Target lift.

The percentage change you expect to drive on your primary success metric as a result of this experiment.

By inviting  connected friends to a workout class, we can improve the conversion rate of class booking by 10%

 

6. Counter metric.

A metric you want to ensure does not suffer at the expense of increasing your primary success metrics.

 

7. Baseline conversion rate.

The current rate of your primary success metrics prior to this experiment.

 

8. Analyzing the experiments.

The experiment was released to the audience selected, and this was the result:

Social experiment increased class attendance by 12.3%.

 

9. Experiment results should be documented.

All your data experiments should be documented with their results in a document.

10. Designing with data.

Here’s how they evolved the app based on data.

Remember. By using data, we are able to: Learn > Build > Measure > Learn > Build > Measure

Does your company already use a data-driven design methodology?

 

 

Avatar
Posted by:Fran Castillo

@francastillo