CSV Analysis for E-commerce: Orders, Products & Customer Insights

Many store platforms let you export orders as a CSV file; check your platform's help pages. Orders, products, customers, refunds: the exports hold a lot. But most store owners never do anything with those exports beyond glancing at totals in a spreadsheet. That is a missed opportunity. Your CSV files contain patterns that standard platform dashboards often do not surface: patterns that can affect profitability, retention, and growth.

Your platform's dashboard and your raw orders can show different things. A Shopify dashboard tells you last month's revenue. A proper analysis of your orders CSV tells you which product combinations drive the highest lifetime value, which customer segments are about to churn, and whether your recent price change actually increased or decreased total margin. This guide walks you through exactly which CSV exports to pull, what analyses to run, and how to turn raw e-commerce data into decisions that move the needle.

Worked Example (Simulated): Sales Rank Is Not Profit Rank

Question: In a made-up set of eight order lines, which of two items earned more gross profit, and is it the one with more sales?

Inputs: Simulated data, seed 23. Each order line has an item (Mug or Tote) and a quantity from 1 to 3. The inputs set the margins: a Mug sells for $20 and costs $11, so it earns $9 a unit. A Tote sells for $24 and costs $20, so it earns $4 a unit.

Method: Total sales and cost of goods by item. Sales are quantity x price. Cost of goods is quantity x unit cost. Gross profit is sales minus cost of goods.

import random

# Simulated data, seed 23. Each row is one order line.
random.seed(23)
products = {
    "Mug": {"price": 20, "unit_cost": 11},
    "Tote": {"price": 24, "unit_cost": 20},
}
orders = []
for order_id in range(1, 9):
    sku = random.choice(["Mug", "Tote"])
    quantity = random.randint(1, 3)
    orders.append((order_id, sku, quantity))

summary = {sku: {"units": 0, "sales": 0, "cost": 0}
           for sku in products}
for order_id, sku, quantity in orders:
    item = products[sku]
    summary[sku]["units"] += quantity
    summary[sku]["sales"] += quantity * item["price"]
    summary[sku]["cost"] += quantity * item["unit_cost"]

print("Simulated data, seed 23")
print("Input: 8 order lines; each has a product and quantity.")
print("Method: sales = quantity x price; "
      "cost of goods = quantity x unit cost;")
print("gross profit = sales - cost of goods.")
for sku, totals in summary.items():
    gross_profit = totals["sales"] - totals["cost"]
    print(f"{sku}: {totals['units']} units, sales ${totals['sales']}, "
          f"cost of goods ${totals['cost']}, gross profit ${gross_profit}")
print("Limit: this small example omits discounts, returns, fees, and shipping.")

Printed output:

Simulated data, seed 23
Input: 8 order lines; each has a product and quantity.
Method: sales = quantity x price; cost of goods = quantity x unit cost;
gross profit = sales - cost of goods.
Mug: 8 units, sales $160, cost of goods $88, gross profit $72
Tote: 8 units, sales $192, cost of goods $160, gross profit $32
Limit: this small example omits discounts, returns, fees, and shipping.

Result: The Tote sold more in dollars ($192 against $160) but earned less gross profit ($32 against $72), because each Tote earns $4 and each Mug $9. The inputs set those margins; the draw gave 8 units of each. So a ranking by sales and a ranking by gross profit can differ.

Limit: These data are made up. Eight order lines are too few to say anything about a real store, and the example omits discounts, returns, fees, and shipping.

E-commerce CSV Data You Should Be Analyzing

Before running any analysis, you need the right data. Every major e-commerce platform exports three core datasets, and each one unlocks different insights.

To see every column of your own file, its type, what is missing and how it is spread, try the free data profile tool.

Orders Export

Your orders CSV is the most valuable file your store produces. It typically contains order ID, order date, customer name or email, line items with quantities, unit prices, discount codes applied, shipping costs, taxes, and order totals. From Shopify, you get this under Orders > Export. WooCommerce uses plugins like WP All Export. Amazon Seller Central provides it under Reports > Fulfillment.

The orders export feeds the majority of high-value analyses: revenue trends, average order value, discount effectiveness, product profitability, and customer purchase patterns. If you only export one file, make it this one.

Products Export

Your products CSV includes SKU, product title, variant details, price, compare-at price, cost per item (if tracked), inventory quantity, and product type or category. This file is essential for profitability analysis. You cannot calculate true margins without cost data. It also enables inventory optimization and pricing experiments.

Customer Export

The customer CSV contains email addresses, total number of orders, total amount spent, location (city, state/province, country), account creation date, and tags. This is the foundation for RFM segmentation and customer analytics, understanding not just what sold, but who bought it and how their behavior changes over time.

Top E-commerce Analyses

The analyses below address questions about products, customers, orders, regions, prices, and repeat buying. Each one answers a specific business question that platform dashboards leave unanswered.

1. Product Profitability

Revenue is vanity, profit is sanity. By joining your orders CSV with your products CSV (matching on SKU or product title), you can calculate true margin per product after accounting for cost of goods, shipping, discounts, and returns. A best-selling product is not always the most profitable one. High-volume items often carry thin margins once discount usage and return rates are factored in.

Product profitability analysis helps you decide where to invest ad spend, which products to bundle, and which items to phase out despite decent sales volume.

2. Customer Segmentation (RFM)

RFM segmentation scores every customer on three dimensions: Recency (when they last ordered), Frequency (how many orders they have placed), and Monetary value (how much they have spent in total). This creates natural segments: Champions who buy often and recently, At-Risk customers whose activity is declining, and lapsed customers who have not purchased in months.

Your customer CSV already contains the fields needed for RFM. Upload it to an analysis tool and you can segment your customer base without building custom reports in your platform.

3. Average Order Value Trends

Tracking AOV over time reveals whether your pricing strategy, upsell tactics, and product mix changes are actually working. A rising AOV with stable order volume means your optimization efforts are paying off. A falling AOV masked by increasing order count is a warning sign. You may be acquiring lower-value customers through discounting.

4. Geographic Sales Patterns

Your orders and customer CSVs contain location data that most store owners ignore. Geographic analysis reveals which regions drive the highest AOV, where return rates are lowest, and whether shipping cost differences create margin disparities by region. These insights inform decisions about regional ad targeting, warehouse placement, and even product assortment by geography.

5. Price Elasticity Testing

If you have changed prices on any product over the past year, your orders CSV holds a before-and-after record. Comparing order volumes and revenue before and after price changes gives a rough, exploratory read on how sensitive customers are to price (see price elasticity analysis). A before-and-after comparison alone cannot separate the price effect from seasonality, promotions, or traffic changes, so treat it as a lead; a comparison group (for example, products whose price did not change) or a proper test gives firmer estimates.

6. Churn Prediction

Customers do not announce they are leaving. They just stop buying. Churn prediction uses order history patterns to flag customers likely to lapse before it happens. Features like declining order frequency, shrinking basket size, and increasing time between purchases feed models that identify at-risk customers with enough lead time to intervene with win-back campaigns.

Platform-Specific CSV Guides

Each e-commerce platform structures its exports slightly differently. Column names vary, date formats differ, and some platforms split data across multiple files where others consolidate.

Shopify provides the cleanest exports with consistent formatting. Our Shopify analytics guide covers the exact export steps and which columns map to each analysis type. For a detailed walkthrough of order-level analysis, see How to Analyze Shopify Orders Export CSV.

WooCommerce requires an export plugin for full flexibility, but the default WordPress export covers orders and products. Column naming follows WordPress conventions, so some field mapping is needed before analysis.

Etsy provides order CSVs under Shop Manager > Settings > Options > Download Data. The export includes transaction fees and shipping labels, making profitability analysis more straightforward than on platforms that exclude fee data.

eBay exports are available through Seller Hub > Orders > Download report. The format includes auction vs. fixed-price indicators, which adds a dimension to pricing analysis that other platforms lack.

Amazon Seller Central offers multiple report types under Reports > Fulfillment and Reports > Business Reports. FBA sellers should pull both the All Orders report and the FBA Fee Preview to capture the full cost picture.

Beyond Basic Reports

Platform dashboards show you what happened. Statistical analysis shows you why it happened and what will happen next. That distinction matters because decisions based on "what" are reactive, while decisions based on "why" are strategic.

Regression analysis estimates links between inputs and revenue. Is it ad spend? Product page views? Email open rates? Discount depth? Regression gives an estimate for each factor's link to revenue. A link is not a cause.

Clustering finds natural groupings in your data that you never defined. While RFM creates segments based on pre-selected dimensions, clustering algorithms like k-means examine all available features simultaneously and discover customer groups that emerge organically from the data. You might find a segment of "weekend bulk buyers" or "holiday-only premium shoppers" that no pre-defined segmentation would have surfaced.

Time-series forecasting projects future demand based on historical order patterns, accounting for seasonality, trends, and external factors. Accurate demand forecasts prevent both stockouts (lost revenue) and overstock (tied-up capital).

These techniques are not just for data scientists anymore. Modern CSV analysis tools run these methods on uploaded files with no coding required. You pick the analysis type, map your columns, and get results.

Start Analyzing Your Store Data

You do not need a data team, a BI tool, or a statistics degree.

Step 1: Export your data. Log into your e-commerce platform and download your orders CSV for the past 12 months. Include all available fields. You can always ignore columns later, but you cannot analyze data you did not export.

Step 2: Upload and analyze. Drop your CSV into MCP Analytics. Select an analysis type: product profitability, RFM segmentation, AOV trends, or let the tool auto-detect the best match for your data structure. The platform handles column mapping, data cleaning, and statistical computation automatically.

Step 3: Act on findings. An analysis can produce findings to review and may help set next steps. Product profitability might tell you to drop a low-margin SKU. RFM might identify 200 at-risk customers who need a win-back email this week. Geographic analysis might reveal an underserved region worth targeting with ads. Pick one finding and execute on it before running the next analysis.

FAQ: E-commerce CSV Analysis