How to measure merchandising effectiveness in an online store?
E-commerce merchandising is not just a nice product order. It decides what customers see first, what gets exposure and which products work for sales, margin and shopping experience.
What does merchandising mean online?
In physical retail, merchandising covers shelves, zones, displays and customer paths. In e-commerce, the equivalents are category order, search results, featured sections, banners, cross-sell, upsell, filters and recommendations.
Every exposure area has limited attention. Products shown higher usually receive more clicks. Merchandising should therefore be measured as seriously as paid campaigns.
Areas that should be measured
| Area | What to measure | Why |
|---|---|---|
| Category list | Position, clicks, conversion and profit. | The main shelf of the online store. |
| Search | Result CTR, no results and exits after search. | Shows whether the store understands intent. |
| Homepage | Clicks, sales after click and margin. | The most expensive exposure space. |
| Recommendations | Attach rate, AOV and add-on sales. | Influences basket value. |
| Banners and internal campaigns | CTR, revenue, profit and stock after campaign. | They can generate traffic without results. |
Merchandising metrics
- Product list CTR: clicks compared with product impressions on a listing.
- Post-click conversion: whether the click ends in a purchase.
- Revenue per view: sales generated by product exposure.
- Profit per view: more important than revenue when profitability matters.
- Share of visibility: how much exposure the product receives compared with its result.
- Basket impact: whether the product increases AOV, cross-sell or add-on sales.
Product order should have a goal
There is no single perfect product order. A category focused on conversion should look different from one focused on stock clearance or margin improvement. The biggest mistake is mixing goals without tracking metrics.
| Goal | Products to expose | Control metric |
|---|---|---|
| Conversion | Products with strong sales effectiveness. | Conversion rate, add-to-cart rate. |
| Margin | High-profit products with reasonable conversion. | Profit per view. |
| Stock turnover | Overstocked products with real demand. | Sell-through and days of inventory. |
| New arrivals | New products with category potential. | CTR, first sales. |
| Basket value | Complementary products and bundles. | AOV and attach rate. |
How to measure a product order change
Treat a listing change as an experiment. Record the date, products moved higher, the goal and baseline metrics. Compare results in a similar period, ideally within the same category and traffic level.
- Choose the category or section to test.
- Define the goal: conversion, margin, stock, new arrivals or AOV.
- Save product positions before the change.
- Apply a data-based ranking.
- Compare CTR, sales, profit and stock after 7-14 days.
- Check whether gains for some products reduced total category performance.
Product position has value
The first positions in a category usually receive the most attention. Treat product position as an asset, similar to ad budget. If a product takes a high position, it should justify it with conversion, profit, basket impact or a strategic goal.
| Position | Risk | What to check |
|---|---|---|
| Top 1-4 | Wasting the most valuable exposure. | CTR, conversion, profit per view, availability. |
| First category page | Low-margin products dominating. | Margin share and assortment diversity. |
| Later pages | Hiding high-potential products. | High-margin products with low exposure. |
| Search results | Poor intent matching. | CTR, no clicks, exits after search. |
Merchandising and on-site search
Search works differently than categories. In a category, users explore; in search, they often have a specific intent. If search results show unavailable, generic or poorly matched products, users quickly lose trust.
- Monitor no-result queries and add synonyms.
- Check queries with many results but low CTR.
- Separate relevance from margin: a profitable product cannot be completely irrelevant.
- Do not let unavailable products dominate top results.
- Analyze searches that lead to exits or no clicks.
A/B test or before/after analysis?
An A/B test is the strongest option, but many small and medium stores do not have the traffic or tools to run it correctly for every category. A disciplined before/after analysis is better than no measurement.
| Method | When to use | Limitation |
|---|---|---|
| A/B test | High traffic, important category, major change. | Requires tooling, time and statistical control. |
| Before/after analysis | Smaller store, quick merchandising test. | Seasonality, campaign or traffic changes can distort results. |
| Category comparison | Similar categories with different sorting logic. | Categories may have different demand and margin. |
| Manual SKU sample test | When testing a selected product group. | Does not show full category impact. |
Common measurement traps
- Looking only at revenue without margin.
- Comparing weeks with different traffic or seasonality.
- Moving products without saving the baseline.
- Promoting products with low stock.
- Hiding niche products that have strong margin and high purchase intent.
- Using the same logic for categories, search and recommendations.
Data needed for a merchandising report
Combine listing, sales and product data: SKU, category, list position, impressions, clicks, units sold, revenue, cost, stock and date of order change. If clicks are not available, start with views, sales, conversion and profit.
Order products by goal, not intuition
Insighteo App helps generate a product ranking from CSV and identify products worth showing higher for sales, profit and conversion.
Create an account and upload CSV