#11 · 2023
NPS Across Quarters for a SaaS Product
Analysis (employer work, de-identified)Self-serve analytics · SaaS
my_role: Individual Contributor (IC) - built the self-serve NPS reporting so teams could track loyalty on their own.
Tracking customer loyalty quarter over quarter. A self-serve NPS reporting setup for a SaaS product, so teams can follow loyalty trends without waiting on an analyst.
## Overview
The analysis is organized to provide two primary forms of output:
- ›Data for export and manipulation - takes in the raw data and creates ready-to-consume data so stakeholders and advanced users can perform ad-hoc analysis for specific use cases
- ›Visualizations that provide a quick overview - so stakeholders and strategic decision-makers can rapidly evaluate key metrics
## Executive summary - descriptive trend analysis
Region C has the highest NPS score and the highest scores for all contributing factors. Region A follows with the second-highest scores, and the remaining regions show the lowest trends for all scores. The overall trend shows a dip in scores at end of 2016 and beginning of 2017; region C, however, follows a distinct pattern not marked by the same seasonality as the other regions.
## Implications
Examine region C further to determine the key contributing factors behind its above-average performance.
## Methods
- ›NPS calculated per quarter and geo in Python/pandas and pivoted (2016Q3–2017Q3) - e.g. Overall Product NPS in region C ranged 58.8–65.5
- ›Colorblind-safe point plots of each factor across quarters
- ›Correlation heatmap across all NPS drivers
- ›OLS regression of Likelihood to Recommend on the drivers (8,135 observations, R² = 0.61): Overall Product (0.55), Realizing Value (0.17), Doing Business With Us (0.14) and Account Management (0.10) were the strongest significant predictors
- year
- 2023
- status
- Analysis (employer work, de-identified)
- tags
- Self-serve analytics