E-commerce sales analysis: what data should an online store really track?
It is easy to focus only on revenue, orders and traffic. That is not enough if you want better decisions about products, promotions, pricing, stock and product visibility.
Revenue does not tell the whole story
High sales can look good in a report, but they do not always mean a healthy business. A product can generate strong revenue while having weak margin, high shipping cost, frequent returns or constant discount pressure.
Sales analysis should answer practical business questions: which products make money, which products only create turnover, which products waste traffic and which products deserve more visibility.
Core data for e-commerce sales analysis
| Data | What it shows | Decision it supports |
|---|---|---|
| Units sold | Real demand for the product. | Whether to keep exposing or reordering it. |
| Revenue | Sales scale and share of turnover. | Which products matter for sales performance. |
| Cost of goods | How much the store pays for the product. | Whether sales are profitable. |
| Gross profit | Revenue minus product cost. | Which products can be promoted without hurting margin. |
| Product views | How much attention a product receives. | Whether visibility is used efficiently. |
| Product conversion | How often views become purchases. | Which products turn traffic into sales. |
| Stock level | Whether the product can continue selling. | Whether to increase visibility, clear stock or reorder. |
| Category | The comparison context. | Whether the product performs well against similar items. |
Metrics worth calculating regularly
- Product conversion: units sold divided by product views.
- Gross profit: revenue minus cost of goods sold.
- Gross margin: gross profit as a percentage of revenue.
- Revenue per view: sales value divided by product views.
- Profit per view: gross profit divided by product views.
- Product share in category sales: useful for spotting leaders and slow movers.
These metrics are simple, but practical. They let you compare products with different prices, traffic levels and margins.
Where should the data come from?
The strongest sales analysis combines several sources. The store platform shows orders and stock, Google Analytics 4 shows user behavior, ad platforms show traffic cost, and accounting or ERP systems help validate costs and margin.
| Source | What it usually contains | What to watch |
|---|---|---|
| Store platform | Orders, products, prices, stock, variants and payment statuses. | Returns, cancellations and discounts may be counted differently than in accounting. |
| GA4 | Product views, add-to-cart events, checkout, purchases and item revenue. | E-commerce data requires events such as `view_item`, `add_to_cart` and `purchase` to be implemented correctly. |
| Ad platform | Campaign cost, clicks, ROAS and conversion cost. | ROAS without margin can push products that look good only by revenue. |
| ERP / accounting | Cost of goods, purchase prices, corrections, invoices and deliveries. | Cost data may arrive late or be available by batch, not by SKU. |
| CSV / spreadsheet | The simplest common format for connecting product data. | Protect SKU identifiers, separators, UTF-8 encoding and consistent column names. |
A minimal dashboard for a store owner
A small or medium-sized store does not need dozens of charts at the beginning. A weekly view that helps decide what to improve, promote and stop reordering is enough.
- Top products by profit: show what actually builds financial performance.
- Top products by revenue: show what drives turnover and operational workload.
- Products with high traffic and low conversion: point to product pages that need improvement.
- Products with high margin and low exposure: are candidates for better category placement or campaigns.
- Products with strong sales and low stock: require replenishment checks before bigger promotions.
- Products with no sales despite stock: move into pricing, image, seasonality or clearance analysis.
This dashboard should be a list of priorities more than a decorative report. The key question is: what will the team do after reading it?
How to read the data without common mistakes
| Signal in data | Possible meaning | What to check next |
|---|---|---|
| Many views, low conversion | The product gets attention but does not persuade customers. | Price, photos, description, availability, reviews and variants. |
| High sales, weak margin | The product creates turnover but may hurt profitability. | Purchase cost, discounts, shipping cost and returns. |
| Strong margin, few views | The product may have potential but low visibility. | Category position, internal links, campaigns and images. |
| Good conversion, low stock | The product works, but stockouts may block sales. | Deliveries, reorder priority and promotion plan. |
| No sales despite stock | The product may be poorly presented, mispriced or mismatched. | Traffic, competitive price and seasonality. |
Segment products because store averages hide problems
Average conversion rate or average margin for the entire store rarely leads to a good decision. Seasonal categories, premium products, accessories and paid-ad bestsellers behave differently.
In practice, analyze products by segments:
- category and subcategory,
- brand or supplier,
- price range,
- main product, accessory, spare part or bundle,
- new arrival, evergreen item, end-of-line product or seasonal product,
- traffic source: organic, paid campaigns, newsletter or marketplace.
Segmentation prevents you from comparing products that should not be judged by one metric. An accessory may have lower revenue but strong margin and basket impact. A premium product may have lower conversion but much higher profit per order.
From report to e-commerce decisions
The best sales analysis ends with a decision. A report does not improve store performance if the team does not know what to do with the data.
- Products with high profit and good conversion are candidates for higher visibility.
- Products with high traffic and weak conversion need better product pages or pricing.
- Low-margin products should be promoted carefully, even if they are bestsellers.
- Products with strong performance and low stock should not be heavily promoted without a replenishment plan.
- Slow-moving products can move into a separate clearance or inventory turnover analysis.
A practical rhythm for sales analysis
Data analysis works best when it has a rhythm. Otherwise reports are opened only after sales drop or inventory becomes a problem.
| Frequency | What to check | Typical decision |
|---|---|---|
| Daily | Sales, payment errors, out-of-stock products and sudden conversion drops. | Operational response and removing sales blockers. |
| Weekly | Top products, products with traffic but no sales, margin, stock and campaigns. | Change visibility, improve product pages and adjust promotions. |
| Monthly | Category profitability, inventory turnover, promotion results and product mix changes. | Purchasing plan, content priorities and clearance decisions. |
| Before season | Sales history, stock, margin, seasonal products and campaign plan. | Reorders, category ranking, bundles and ad budgets. |
How to start if you only have a CSV file
You do not need a data warehouse or a complex BI setup to begin. A practical first report can be built from a CSV export if it contains product ID, name, category, units sold, views, revenue, cost and stock level.
That file is enough to create a product ranking by profit, conversion and visibility potential. It is a good starting point for store owners, e-commerce managers, performance marketers and merchandising teams.
Data quality checklist before analysis
- Every product has a stable SKU or product ID.
- Variants are separated when they have separate stock, price or sales.
- Revenue, cost, units sold and views cover the same time period.
- Returns, discounts and cancellations are handled consistently.
- Categories are not empty and do not change names between exports.
- Numbers are stored as numbers, without labels such as currency, units or comments.
- The file uses UTF-8 so special characters in product and category names do not break analysis.
Turn product data into priorities
Insighteo App lets you upload a CSV file, map columns and see which products should be shown higher, improved or reviewed for profitability.
Create an account and upload CSV