Quick answer: A session is one visitor's trip to your listings inside a 24-hour window, counted once. A page view is every single load of a detail page. Amazon's conversion metric divides orders by sessions, so that's the traffic number that matters for judging a listing.
Open the Sales and Traffic report and you get both numbers side by side, usually with page views comfortably larger than sessions. It's tempting to treat them as roughly the same signal at different resolutions. They aren't, and the gap between them carries information most sellers skip past.
How Amazon defines each one
A session is a period of activity by a single user on your product pages, within a 24-hour window. If someone lands on your listing, clicks around, leaves, and comes back four hours later, that's still one session. Reload the page ten times in a row and it's one session. The count only ticks up again when a genuinely separate visit happens, or when the 24-hour clock rolls over. Amazon's Business Reports documentation describes it in those terms, and the practical effect is that sessions approximate how many shopping trips your catalog got.
A page view is exactly what it sounds like: one load of one detail page. No deduplication. The same shopper bouncing between your parent listing and three child variations generates four page views inside a single session. Someone who opens your product in two browser tabs to compare it against a competitor doubles the page views without adding a session.
So the ratio of page views to sessions is roughly pages per visit. A listing with lots of variations, or a comparison-heavy category, will naturally run a higher ratio than a single-SKU consumable that people buy on autopilot.
Why conversion is calculated on sessions, not page views
Amazon's conversion metric is Unit Session Percentage: units ordered divided by sessions. It is not units divided by page views, and the choice is deliberate. Conversion is meant to answer "of the people who showed up to shop, how many bought?" — and the honest denominator for that question is visits, not page loads.
If you accidentally reason about conversion using page views, every listing looks worse than it is, and variation-heavy listings look worst of all, purely because their shoppers click through more child ASINs per visit. That's an artifact of catalog structure, not a demand problem. We walk through the session-based version of this in tracking conversion rate by ASIN, and the category-level context in Amazon conversion rate benchmarks.
The traffic-quality read hiding in the ratio
Because page views per session reflects how much comparison shopping is happening, a sudden move in that ratio is worth a look. A listing that jumps from 1.4 page views per session to 2.6 without a catalog change often means shoppers arriving with weaker intent — broad ad terms, an off-target deal placement, a new competitor pulling people into a side-by-side. They're browsing more and buying less, and conversion usually softens a few days later.
One seller noticed page views climbing week over week while sessions stayed flat and orders slipped. The cause turned out to be an automatic ad campaign that had drifted onto a cluster of loosely related search terms. The traffic looked fine in the headline session count; the quality was visible only in how many pages those visitors churned through before leaving. Pausing the drifted terms brought the ratio and the conversion rate back within about a week.
The B2B split most reports bury
Both metrics come in a "Total" flavor and a "B2B" flavor. Sessions – B2B and Page Views – B2B count only signed-in Amazon Business buyers. If you sell anything with business appeal — office supplies, tools, bulk packs — the B2B slice can be a meaningful share of sessions and often converts at a different rate than consumer traffic. Judging a listing on the blended number alone can hide a B2B conversion problem, or a B2B opportunity, underneath a consumer average that looks unremarkable. This is the same averaging trap that lets an account-wide figure hide a struggling ASIN.
What to actually do with the two numbers
Use sessions as your traffic denominator for anything conversion-related, and as the weight when you roll listings up — a 40% conversion drop on a 20-session ASIN is noise, the same drop on a 600-session ASIN is an emergency. Use page views per session as a secondary gauge of intent: stable is fine, a sharp rise alongside flat sessions and falling orders points at where your traffic is coming from, not at the listing itself. And check the B2B columns on anything a business might buy, because the blended view treats those two audiences as one.
Neither number updates in real time. Both lag a day or more and can restate as Amazon finalizes the data, which is worth remembering before you react to a single day's move — see how often Seller Central updates.
EcomSanity's Buy Box and conversion widget works in session-weighted terms per ASIN, so a listing losing conversion on real traffic surfaces ahead of low-volume noise. If you want the full field guide to the report these numbers come from, see how to read Amazon's Sales and Traffic report.
Frequently asked questions
What's the difference between sessions and page views on Amazon?
A session is one visit by one user to your product pages within a 24-hour window, counted once no matter how many times they reload or come back that day. A page view is every individual load of a detail page, so one session usually contains several page views.
Which one does Amazon use to calculate conversion?
Sessions. Unit Session Percentage is units ordered divided by sessions, not by page views, which is why it's the number that actually tracks how well a listing turns visitors into buyers.
Why is my page views number so much higher than sessions?
Because page views aren't deduplicated. A shopper who opens your listing, clicks to a variation, comes back, and checks it again the next morning is one or two sessions but four or five page views.