Improved examination of the influence of product attributes on customer retention
=================================================================
In the realm of product analysis, understanding the retention of features is crucial. A recent study has shed light on the distribution of product features based on usage and retention. Here's how a well-designed scatter plot can help visualise this data.
The scatter plot in question combines two essential metrics: the percentage of returned users (a qualitative metric) and the number of users (a quantitative metric). To ensure clarity, follow these guidelines:
- Appropriate Axis Assignments: Place the truly quantitative variable (number of users) on the X-axis and the percentage returned users (a ratio/percentage, which can be treated as a continuous variable) on the Y-axis.
- Color Coding or Distinct Markers: If the "qualitative" aspect refers to categories or user segments within returned users, differentiate these groups using different colors or symbols to add clarity and enable visual separation of segments.
- Clear, Descriptive Title, Axis Labels, and Legends: This helps viewers immediately understand what each axis represents and what the colors or markers signify.
- Add Trendlines or Smoothing Lines: To show the overall relationship or trend between percent returned users and number of users, a trendline can clarify patterns in retention effectiveness.
- Highlight Outliers or Notable Points: For example, if certain features have exceptionally high retention but low user count, size the points differently or annotate these to draw attention.
- Simplify and Declutter: Remove unnecessary gridlines, limit color palette to a few distinguishable hues, and keep labels concise to reduce cognitive load.
- Consider Additional Visual Encoding If Needed: Sometimes adding point size to represent a third variable or adding small multiple plots (facets) can help interpret complex multi-metric data.
After applying 50/80 percentile rules, 9 clusters of product features were identified in the scatter plot. The X-axis measures the % users, representing the popularity of a product feature, while the Y-axis measures the % returned users, representing the value of a product feature.
The first two product features with the highest % returned users have a modest volume in terms of users. These are the "Niche" features, a special subset that brings enormous value to a limited number of users (11.3%).
On the other hand, "Power2" features bring as much value as core features but are less popular. These features have a percentage of users within the 50-80th percentile and a high percentage of returned users.
The "Core" features, a small number (2-3), are used by more than 80% of users and have a high percentage of returned users.
"Casual1" features are used from time to time, with a usage percentage within the 50-80th percentile and a usage percentage of returned users within the same range.
Lastly, there's the "Set-up" features, a unique subset designed to set up a product for convenient usage. These features are used by a lot of users, but this usage typically happens only once, during the onboarding phase.
This problem is common in many analyses, where a quality measure is presented without a quantity measure. Combining both qualitative and quantitative metrics into one chart can help fix this issue, providing a more comprehensive understanding of product feature retention.
Technology plays a pivotal role in visualizing product feature retention, with data-and-cloud computing facilitating the creation and analysis of scatter plots. This data representation, when combined effectively with technology, reveals valuable insights about the performance of different product features.
Scatter plots specifically, propelled by technology, enable the simultaneous observation of quantitative and qualitative metrics, thereby offering a deeper understanding of product usage patterns and retention rates.
