Where the test’s visitors click, how far they read, and what stays on their screen the longest, variant by variant, with only the differences that hold between variants. The maps are built from session recordings. They never include the visitor’s text, images, or input, and nothing beyond what session recordings already capture.
maisonlune.fr/panierDemo store page. All devices, shown at the most common width (mobile).
Elements clicked and depths reached on each side, all devices. Only differences that hold are flagged.
Share of the page’s visitors who click the element at least once, or whose screen reaches that depth, Original vs. Progress bar. 19 comparisons (14 elements, 5 depths): differences that hold are selected so that, on average, no more than 5% of the flagged ones are false, and each range, at 97.6%, is widened accordingly, so it excludes zero exactly when a difference holds. A difference in clicks describes behavior. It doesn’t say which variant is better: only the test’s metric decides that.
Each click is recorded on the element clicked, identified by a stable selector. A burst of clicks on the same element counts as one click, and becomes a rage click at three clicks within 700 ms. A dead click changed nothing on the page within a second. An element’s color follows its share of the page’s clicks, on the same scale for every variant; the dots show where people click inside the element.
Scroll depth is how far down the bottom of the screen reaches, as a share of the page height, keeping each visitor’s deepest point. Attention is the screen time spent on each tenth of the page, derived from continuously recorded scrolling. A variant gets its map from 30 visitors on the page. An element or a depth is compared once each variant has 5 visitors who click it (or reach it) and as many who don’t.
There are many comparisons (every element, every depth), so chance alone would make a few stand out. Flagged differences are selected so that, on average, no more than 5% of them are false, and their ranges are widened accordingly.
No input or field values. Inside a form field or an area marked data-probant-mask, neither the label nor the click position. Nothing before the visitor’s consent when the workspace requires it. Session recordings, and therefore maps, are kept for 30 days. The preview is the public page loaded when you view it, without scripts or cookies. The masked page structure, when the workspace records it, is used only for replay: never the visitor’s text, images, or input.