Business Analysis

Customer Analysis Example

A customer-behavior analysis using a U.S. transaction dataset to identify differences in value, purchasing frequency and recent activity.

United StatesUSDJanuary 2024 – December 2025
BUSINESS DATA ANALYSIS
Project Resources

Why the analysis was needed

A business had a growing customer base and a large transaction history, but customer value and activity varied considerably. Management wanted a clearer way to distinguish repeat, new, occasional and less-active customers using consistent customer-level measures.

BUSINESS QUESTION

Which customer groups contribute the most value, how does purchasing behavior differ between groups, and where are there meaningful differences in recent activity?

What the analysis used

Customer and transaction records covering 8,240 customers and 38,460 transactions.

RECORDS38,460 transactions
CUSTOMERS8,240 customers
FIELDSCustomer ID, transaction date, transaction value, product category, acquisition channel, order count and last purchase date
PERIODJanuary 2024 – December 2025

How the source was made analysis-ready

  1. 01

    Validated customer identifiers, transaction dates and transaction values before customer-level aggregation.

  2. 02

    Aggregated transaction history to one analytical record per customer while preserving transaction counts and total historical value.

  3. 03

    Calculated recency from the end of the analysis period, purchase frequency from transaction counts and monetary value from transaction totals.

  4. 04

    Created mutually exclusive behavioral segments so every customer is represented once in the segment summaries.

How the analysis was structured

  1. 01

    Aggregated transactions to the customer level.

  2. 02

    Calculated recency, purchase frequency and customer monetary value.

  3. 03

    Grouped customers into business-friendly behavioral segments.

  4. 04

    Compared customer counts and revenue contribution across segments.

  5. 05

    Reviewed the distribution of value and activity to identify materially different customer groups.

Customers8,240Unique customers analyzed
Transactions38,460Transactions in the analysis period
Revenue$2.740MTransaction value across the dataset
High-Value Revenue Share27.0%Revenue from 660 high-value repeat customers
Customers by SegmentDistribution of the customer base
Customers
High-value repeat660
Regular1,920
New1,720
Occasional2,130
At-risk1,080
Inactive730
Revenue by SegmentRevenue contribution by customer group
$000s
High-value repeat$740K
Regular$760K
New$320K
Occasional$390K
At-risk$340K
Inactive$190K
01

A small group contributes a large share of revenue

High-value repeat customers represented 660 customers, or 8.0% of the customer base, but generated $740K, approximately 27.0% of total revenue.

02

Regular customers form the largest revenue pool

The Regular segment generated $760K from 1,920 customers, slightly exceeding the revenue contribution of the High-value repeat segment.

03

Customer activity varies substantially

The analysis separates recent and repeat behavior from customers with lower frequency or longer gaps since their last purchase, making the customer base easier to prioritize for further analysis.

04

At-risk customers represent a distinct analytical group

The At-risk segment contains 1,080 customers and $340K in historical revenue, making it materially different from the Inactive segment and useful for retention-focused analysis.

Customer Segment Summary
SegmentCustomersRevenueAverage Revenue / Customer
High-value repeat660$740K$1,121
Regular1,920$760K$396
New1,720$320K$186
Occasional2,130$390K$183
At-risk1,080$340K$315
Inactive730$190K$260

What the numbers mean

The segmentation shows why customer count alone is not enough to understand the customer base. Revenue contribution, frequency and recency create materially different groups that can be analyzed separately for retention, CRM and customer-value planning.

What was checked before interpretation

  • Segment customer counts reconcile to all 8,240 customers in the prepared customer table.
  • Segment revenue reconciles to the $2.740M transaction-value total.
  • The 660 High-value repeat customers represent 8.0% of customers and their $740K represents approximately 27.0% of revenue.
  • Average revenue per customer in the supporting table was recalculated from each segment's revenue and customer count.

How the findings could inform the next decision

01

Protect high-value repeat relationships

Use the High-value repeat group as a priority population for deeper retention and product-mix analysis because 8.0% of customers account for about 27.0% of historical revenue.

02

Separate at-risk from inactive customers

Test retention questions on the 1,080 At-risk customers separately from the Inactive group rather than treating all less-recent customers as one population.

03

Track movement between segments

For recurring analysis, measure how customers move between New, Regular, At-risk and other states over time instead of treating the segmentation as a one-time label.

What this example does not establish

  • Customer segments are defined using behavioral characteristics and clearly specified thresholds, providing a consistent basis for comparison and interpretation.
  • Historical transaction value is not the same as customer lifetime value because future purchases, servicing cost and margin are not modeled.
  • The example does not include campaign exposure or experimental evidence, so it cannot show that a retention action caused a customer to purchase.
  • Acquisition channel is available in the source profile but channel economics are outside this focused customer-segmentation example.

DELIVERABLE

What a comparable engagement could produce

01Customer segmentation
02Customer value and activity analysis
03Segment comparison table
04Customer behavior findings
05Potential CRM analysis framework

Why the output matters

The output gives management a clearer customer-level view and a structured basis for deciding which groups warrant deeper retention, acquisition or lifecycle analysis.

NEXT

Have a similar business question?