Use Cases

May 16, 2024

How American Express uses Sales Analytics to issue over 115 million credit cards to their customers

Wanna know how a leading full-service commercial bank like HDFC Bank might be using customer demographic & product fit analysis method to improve financial performance? Read here! :)

American Express is a big name in credit cards, with over 115 million cards around the world. They focus on high-end credit cards and compete with companies like Chase and Bank of America. In the U.S. alone, they've issued over 55 million cards, for everyone from individuals to big businesses. Each year, people spend more than $1.6 trillion using American Express cards, mostly online through their website and app.

Security is a major priority for American Express. They use advanced technology to fight fraud, and less than 0.01% of transactions turn out to be fraudulent. This focus on security helps keep customers happy, with satisfaction scores often exceeding 90% in important markets.

One of the biggest perks of American Express cards is their rewards program. Through Membership Rewards, cardholders earn points worth over $7 billion every year. These points can be redeemed for things like travel, gift cards, or even paying your bill, giving cardholders valuable rewards for their spending.

Try this Business Use Case by yourself here👇
Slideshow Demo
Product Fit Analysis Data Story

About Dataset

Let’s Inspect our Credit cards sales Data. The dataset consists of six entries capturing credit card transactions, each delineating key attributes such as customer details, transaction specifics, and financial terms. It includes fields like customer name, advisor handling the transaction, annual fee, commission earned, transaction category (predominantly "CREDIT CARDS"), and transaction dates. Additionally, managerial hierarchies, including national heads, zone heads, and sales managers, are documented alongside location specifics such as city, state, and PIN code. The dataset provides insights into credit card transactions facilitated by advisors, detailing associated fees, commissions, and payment cycles, as well as the specific credit card products and banking institutions involved.

Vehicle Insurance Dataset Fields
Column Name Data Type Unique Values Description
row_id Integer Unique identifier for each row.
lead_code String 15029 Unique code associated with each lead.
customer_name String 11784 Name of the customer.
advisor String Name of the advisor.
advisor_code String 196 Unique code associated with each advisor.
city String 77 Customer's city.
state String 17 Customer's state.
pincode Integer 137 Postal code for the customer's location.
annual_fee Float 29 The annual fee associated with the credit card.
commission Float 34 Commission amount for the sale.
commission_cycle Integer The cycle or period for commission payment.
category String 2 The category of the product, in this case, "CREDIT CARDS".
credit_card_name String 32 Name of the credit card.
bank String 10 The bank issuing the credit card.
national_head String 2 Name of the national head.
zone_head String 4 Name of the zone head.
sales_manager String 20 Name of the sales manager.
dt_payin Date The date when payment was made/received.
dt_payout Date The date when payout (commission) was made.

The Customer Loan Data contains over 15712 rows and 12 of our Columns have String Data type, 3 of our Columns have Integers Data Type, 2 column have Float Data type and 2 of the columns have Date Data Type.

Dataset Here

How American Express uses its Analytics ?

To understand the approach for this kind of analysis, we need to consider these four crucial elements:

  • Target Audience: Identify the primary users of this analysis, such as managers or executives.
  • User Needs and Challenges: Understand the objectives and challenges of these users. What do they hope to achieve? What obstacles are they facing?
  • Key Performance Indicators (KPIs): Determine the most important metrics for measuring advisor performance.
  • Analysis Objectives: Define the overall goals of this analysis. What specific insights are we aiming to uncover?

Step 1: Understanding the Business Persona 

The Sales Manager is the leader of the sales team, the one responsible for steering them towards achieving the company's sales goals. They act as strategist, motivator, and coach. They craft sales plans, identify new business opportunities, and monitor overall sales performance. They collaborate with marketing and product teams to create winning sales materials and products. 

User Persona: Sales Manager

  • Keeping an eye on sales plans, ensuring they align with company goals and are on track for success.
  • Monitoring sales advisors and their strategies.
  • Overseeing overall sales performance and making sure the team hits their targets.
  • Coming up with new strategies and understanding the market.
  • Segregating the market according to the performance.

  • Coming up with new strategies for every state market.
  • Keeping customers satisfied.
  • Head-on for any new Challenge thrown towards them.
  • Handling the low effort revenue state and cities.
  • Generating new leads.

Step 2: Design Empathy Map

To understand a sales manager better, we use an empathy map. It helps us see their feelings, goals, and problems more clearly. This way, we can create a data story that not only works well but also connects with their feelings and needs.

Step 3: Identify the Key Performance Indicators (KPI’s)

After we build this empathy map, we need to pick the most important things to watch. These are basically the key points that tell us how well things are going, kind of like a progress report. We call them KPIs, which stands for Key Performance Indicators. These KPIs help us understand what's working for our customers and products.

Vehicle Insurance Dataset Fields
KPI Formula Description
Credit Card Sales by City Count of lead_code per city Measures the total number of credit cards sold in each city, highlighting regional preferences and product market fit.
Average Annual Fee by State Sum of annual_fee by state / Count of lead_code by state Calculates the average annual fee customers are charged for credit cards in each state, indicating economic capabilities and preferences at a regional level.
Commission Earned per Advisor Sum of commission by advisor Evaluates the total commission earned by each advisor, reflecting on their performance and the product's profitability to them.
Bank-wise Credit Card Sales Count of lead_code per bank Identifies the volume of credit card sales associated with each bank, highlighting bank popularity and the success of their products in the market.
Top Performing Credit Card Category Count of lead_code per category Determines the most popular credit card category (e.g., CREDIT CARDS), showing customer preference for specific card benefits or features.
State-wise Distribution of Commission Sum of commission by state Shows how commission earnings are distributed across states, providing insights into regional market performance and the effectiveness of advisors in those areas.
Credit Card Name Impact on Sales Count of lead_code per credit_card_name Reveals the impact of specific credit card names on sales volume, indicating product popularity and market acceptance of credit card features offered by the bank.

Step 4: Understand the Goals & Objectives of User

Now that we know what's important to the sales managers, we need to set some goals. These goals will help us make decisions based on the data we have. Here's what we want to achieve:

  • Leverage data & KPIs for insights.
  • Optimize sales strategies based on data.
  • Foster collaboration across sales, marketing & product teams.
  • Achieve industry leadership through these actions.

  • Increase credit card sales revenue within the next economic year, positioning the company as a top performer in the industry.
  • Improve conversion rates  through targeted sales strategies informed by comprehensive sales data analytics and customer feedback.
  • Enhance sales team productivity and effectiveness by implementing tailored training programs and providing access to cutting-edge sales tools and technologies.
  • Identify and capitalize on emerging market trends and customer preferences to gain a competitive edge and expand market share.

Step 5 : Ask Business Questions

KPIs can tell you what's happening, but strategic business questions are crucial for understanding why it's happening and how it aligns with your overall business goals. Therefore, here are some business questions to ask to your dataset.

Q1. What is the number of advisors operating in 2023 and how much commissions have  they earned?
  • Metric: Number of Advisors, Total Commissions
  • Question:
    • Unique Advisors in 2023
    • Commission in 2023
  • Observation: In 2023, 232 advisors earned commissions of a total 8.3 Millions.

Q2. Who are the top 10 advisors by total commissions earned in 2023?
  • Metric: Total Commission,
  • Question:
    • Commission by top 10 advisor

Q3. What are the top performing advisor states in 2023?
  • Metric: Total Customer count
  • Question:
    • Top 5 states by count of customer in 2023
    • Top 5 cities by count of customer in 2023
  • Observation: Delhi and Rajasthan have maximum number of customers followed by UP, Maharashtra and West Bengal respectively, while in terms of cities Jaipur, Delhi, Mumbai lead the chart.

Q4. How effective are advisors in their sales performance? Is there a correlation between the total customer count and the average annual fee?
  • Metric: Customer count, Average annual fee
  • Observation: Analysis of 2023 data reveals that the majority of advisors are situated in a lower performing category for both customer count and average annual fee. Prominently, top-performing advisors such as Virendra Singh Shekhawat and Divya Keshagoni demonstrate a specific customer base but lower average annual fees.

Q5. Is there any correlation between average commission cycle and total number of customers?
  • Metric: Average Commission Cycle, Customer Count
  • Question:
    • Advisor by average commission cycle and customer count
  • Observation: We can observe that most of the advisors operate between 15 - 20 days of commission cycle. Other than this, the number of customers is significantly low. Only 4 advisors have a high count of customers with 15-20 days of cycle.


  • Strong Overall Performance: In 2023, our team of 232 advisors generated a total of $8.3 million in commissions, demonstrating solid sales performance.
  • Top Performers Shine: Prominently, the top 10 advisors within the team achieved significant success, highlighting their exceptional sales capabilities.
  • Geographic Strength: Delhi and Rajasthan emerged as the leading regions in terms of customer base, followed by Uttar Pradesh, Maharashtra, and West Bengal.
  • Top Cities: Jaipur, Delhi, and Mumbai stood out as the cities with the highest concentration of customers


  • This way we get to know our advisors better! We can look at how many new customers each advisor brought in last year (2023) and the total annual fees collected from those customers. This will help us see who's building a strong client base and generating revenue.
  • Looking at advisors by region showed us exactly where we're excelling, like Delhi and Rajasthan, and where there's place for improvement, like Uttar Pradesh and West Bengal. This way, we can focus on giving extra support to those areas and help them become customer magnets too.
  • Lastly, we also get to see how advisors work, we can do this by looking at the average annual fee, the commissions they earn, and the number of customers they bring in all play together. This will give us clues about their strategies - are they focusing on high-value clients with bigger fees, or building a larger base with smaller fees?
Ready to get started?

Join Data Analysts who use Super AI to build world‑class real‑time data experiences.

Request Early Access