Net Promoter Score Quadrant Analysis
Visualizing Customer Ratings and Expectations, on a Quadrant Plot using matplotlib in Python
Published on by Vimal Octavius PJ
Quadrant Plot NPS
224 min READ
Although the data is fictitious, the approach and visualizations are pretty much from the projects I've worked on, directly. The actionable outcomes were followed through to fruition.
The NPS data contains the following fields:
- Category Column contains the areas of focus the customer has rated the service
- NPS_Mean The Average of Net Promoter Score given by various Customers, collated by Category so it appears against the corresponding Category
- Score_Count Count of respondents for each category
- Percent_Contribution The individual Contribution of that specific Category to the total, expressed as a percentage (.169 translates to 16.9%, .158 is 15.8% and so on...)
Creating a Scatter Plot with quadrants¶
import matplotlib.pyplot as plt
%matplotlib inline
import warnings
warnings.filterwarnings('ignore')
import pandas as pd
nps_data = pd.read_csv('/content/drive/MyDrive/Datasets/NPS-Prediction - Summary.csv')
nps_data.head()
| Category | NPS_Mean | Score_Count | Percent_Contribution | |
|---|---|---|---|---|
| 0 | OpEx | 8.500000 | 104 | 0.169657 |
| 1 | Resource Quality | 7.800000 | 97 | 0.158238 |
| 2 | Continuous Improvement | 8.000000 | 75 | 0.122349 |
| 3 | Cost Efficiency | 7.666667 | 66 | 0.107667 |
| 4 | Leadership | 7.808511 | 47 | 0.076672 |
import matplotlib.pyplot as plt
from matplotlib.pyplot import figure
%matplotlib inline
def scatter_plot():
fig1, ax1 = plt.subplots(figsize=(11,8))
ax1.scatter(x=nps_data['NPS_Mean'],y=nps_data['Percent_Contribution'])
ax1.set_xlabel('Average Net Promoter Score')
ax1.set_ylabel('Category % Contribution')
ax1.set_title('Key Customer Expectations')
plt.plot([7.5,7.5],[0,.16], linewidth=1, color='red')
plt.axhline(y=0.05, color='r', linestyle='-')
for idx, row in nps_data.iterrows():
ax1.annotate(row['Category'], (row['NPS_Mean'], row['Percent_Contribution']))
plt.show()
with plt.style.context('dark_background'):
scatter_plot()
Takeaways from the above visualization:¶
Splitting the scatter plot into 4 quadrants, breaking up data points that fall before and after 7.5 Average Net Promoter Score (X axis) and, above and below a 5% Contribution of various Categories (Y axis) our Customers have rated the Service on.
- Top 3 aspects to continue the good work on:
- Operational Excellence (OpEx)
- Quality of Resources Provided and,
- Continue to do well on Continuous Improvement Projects
- Top Improvement Areas could be:
- Providing Partnership - work in partnership with the Customer on solutions provided, and,
- Be Proactive
Tracking Movement of Net Promoter Score¶
- Considering "GREEN" NPS Score or "Promoters" as those Customers who've given a rating of 9 or 10 (On a Scale of 1-10)
- "RED" as detractors, Customers who rate the service or product as 3 or below, let's draw a Quadrant like below:
font_darkgreen = {'family': 'serif',
'color': 'darkolivegreen',
'weight': 'bold',
'size': 13,
}
font_green = {'family': 'serif',
'color': 'green',
'weight': 'bold',
'size': 13,
}
font_darkred = {'family': 'serif',
'color': 'darkred',
'weight': 'bold',
'size': 13,
}
font_red = {'family': 'serif',
'color': 'red',
'weight': 'bold',
'size': 13,
}
font_blue = {'family': 'serif',
'color': 'midnightblue',
'weight': 'bold',
'size': 16,
}
plt.figure()
plt.xlim((1,9))
plt.ylim((1,9))
# Drawing Lines
plt.plot([5,5],[1,9], linewidth=3, color='blue' )
plt.plot([1,9],[5,5], linewidth=3, color='blue' )
plt.title('Net Promoter Score - Quadrant Analysis ',font_blue)
plt.rcParams["figure.figsize"] = (10,10)
#Quadrant Title
plt.text(5.2,8.5,"(A) NPS Green in the Last 2 Years",font_darkgreen)
plt.text(1.2,8.5,"(B) NPS Turned to Green from Red",font_green)
plt.text(1.2,1.5,"(C) NPS Turned Red from Green",font_red)
plt.text(5.2,1.5,"(D) NPS Red in past 2 Years",font_darkred)
#Quadrant Customer Names
plt.text(5.5,8,"(1) Customer - A")
plt.text(5.5,7.5,"(2) Customer - B")
plt.text(5.5,7,"(3) Customer - C")
plt.text(5.5,6.5,"(4) Customer - D")
plt.text(5.5,6,"(5) Customer - E")
plt.text(1.5,8,"(1) Customer - F")
plt.text(1.5,7.5,"(2) Customer - G")
plt.text(1.5,7,"(3) Customer - H")
plt.text(1.5,4,"(1) Customer - I")
plt.text(1.5,3.5,"(2) Customer - J")
plt.text(1.5,3,"(3) Customer - K")
plt.text(5.5,4,"(1) Customer - L")
plt.text(5.5,3.5,"(2) Customer - M")
#Remove X and Y ticks - Hide coordinates
plt.xticks([])
plt.yticks([])
plt.show()
The above quadrant aims to track the NPS over the past 2 years. A few examples of actions that can be taken based on the quadrant are:¶
Aim to sustain and build stronger relationships with customers in Quadrant A
For Customers in Quadrant B, work on building stronger improvement plans, followed by execution to help move these Customers to Quadrant A
Customers in Quadrant C need actions to pull back the score to Green
Customers in Quadrant D (NPS RED in the past 2 years) need targeted break-through improvements