Unit Session Percentage: what it measures and what it misses
Unit Session Percentage is commonly called Amazon conversion rate, but it measures units per session. Learn how to calculate it and read it correctly.
Unit Session Percentage is usually introduced as Amazon's version of conversion rate. That shortcut is useful until shoppers begin ordering more than one unit.
The metric counts units, not buyers and not orders.
The formula
Unit Session Percentage = units ordered / sessions x 100
Suppose an ASIN receives 100 sessions. Customers place eight order items and purchase 12 units in total.
- Order-item conversion is roughly 8 percent
- Unit Session Percentage is 12 percent
The listing did not convert 12 separate shoppers. It produced 12 units from 100 sessions.
For a product frequently purchased in multiples, that distinction can be large. Unit Session Percentage can even exceed 100 percent when the number of units ordered is greater than the number of sessions.
What the metric tells you
Unit Session Percentage measures the unit yield of product-page traffic.
When it improves, one or both of these things happened:
- A greater share of sessions produced a purchase
- Buyers ordered more units per purchase
Both are commercially useful. They are different behaviors.
That is why Unit Session Percentage is most helpful beside total order items and average units per order item. Those metrics show whether the change came from more buying sessions or larger baskets.
Amazon includes Unit Session Percentage in its Sales and Traffic reports by date and ASIN. Amazon's report schema shows it alongside sessions, order-item session percentage, and Featured Offer percentage.
What it does not tell you
Whether the traffic was good
A listing can maintain its usual Unit Session Percentage while sessions collapse. Conversion looks healthy, but the product is reaching far fewer shoppers.
The reverse can happen after a broad advertising push. Sessions rise quickly, units rise more slowly, and Unit Session Percentage falls. That does not automatically make the campaign a failure. The product may still be selling more units profitably.
Whether you owned the Featured Offer
Sessions belong to the product detail page. Your offer may not have been the primary offer throughout those visits. A lower Featured Offer percentage can pull down your units per session even when the page itself is attracting the same shoppers.
Whether the sales were profitable
Discounting can improve Unit Session Percentage while reducing contribution margin. The metric knows how many units moved. It knows nothing about what you earned on them.
How many people bought
Multiple units in one order raise Unit Session Percentage. Use order items or order-level data when the question is how many purchases occurred.
How to read it without fooling yourself
First, keep the ASIN and time period consistent. Comparing a parent family this month with one child last month mixes two different populations.
Second, look at volume. Moving from one unit to two units on ten sessions doubles the percentage, but it is weak evidence of a lasting change.
Third, check the surrounding metrics:
| Unit Session Percentage | Sessions | What to ask next |
|---|---|---|
| Down | Flat | Did the offer, price, delivery promise, reviews, or listing change? |
| Flat | Down | Why did traffic or demand fall? |
| Down | Up | Did a new traffic source bring less-qualified shoppers? |
| Up | Flat | Did more sessions buy, or did buyers order more units? |
| Up | Down | Did the remaining traffic become more concentrated or more qualified? |
What is a good Unit Session Percentage?
There is no useful universal answer. Product price, category, replenishment behavior, pack size, brand strength, traffic source, and lifecycle all change the baseline.
Build the benchmark from the product's own history. Compare the same child ASIN across similar periods, then compare it with genuinely similar products in your catalog. A premium appliance and a low-priced consumable should not share a target.
Unit Session Percentage is a strong diagnostic metric once you call it what it is: units produced per session. Treating it as a complete account of conversion hides the behavior that made it move.