A session is not something the web does. HTTP requests do not arrive with a label that says "this is still the same visit." A product invents the visit by deciding which requests belong together. The usual rule is: same browser, and the requests are close enough in time.
The common rules
Universal Analytics popularized a 30-minute timeout. A hit, then silence, then a hit 31 minutes later, was a new session. A new day's midnight could also cut a session. A change of campaign tags could cut one too, so one person clicking two of your links became two sessions. GA4 kept a timeout and changed other details. Other products pick 15 minutes, or a day, or they never build sessions at all.
That is why "sessions" from two tools disagree even when both saw the same people. They are counting different groupings. Bounce rate only makes sense inside one of those groupings: a session that never produced a second hit.
What MapMyVisitors does instead
The widget records loads, with a time, a place, a page, and a referrer. It does not glue those loads into a visit, and it does not publish sessions or bounce rate. If you need "how many visits, where a visit is a sitting," use a product that defines a sitting. If you need "is this site being fetched, and from which countries," the load list is the right unit and a session would only hide it.
A pageview is the single load. A session is zero or more of those, packaged. Quoting sessions from one tool and pageviews from another divides numbers that were never in the same set.
How to read the word in a report
Ask which timeout, and whether a new campaign starts a new session. If the report does not say, do not compare it with last year's number from a different product. The glossary entry session is the short form. The date range you choose still matters more than the timeout: a month of sessions and a day of sessions are not the same idea scaled up. Choosing a date range comes before arguing about the 30 minutes.
