September 22, 2021
Your cancel flow is the opportunity to uncover the specific problem that is driving your customer to cancel their subscription. If you know the specific problem it is easier to overcome it with the optimal solution for that problem. However, there's an issue lurking in how you ask the question.
In your cancel flow, when asking "Why do you want to cancel" (or some variation) you might be injecting bias into the results.
There are two common issues that can appear in choice-based forms:
1. Primacy Bias:
When customers select the first option they see. This happens with customers who have not taken the time to read through the choices.
2. Recency Bias:
When customers select the last option they see. This happens when customers choose the last option as it is the most memorable.
The simplest solution for removing bias in your cancel flow is to implement randomization. Randomization will switch the order of the choices continuously, thereby eliminating the broader impact of primacy and recency bias.
A simple change like this will make your cancel flow more effective and capture more accurate data.
Note: RetentionEngine performs randomization by default.