Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

Glossary term

Session Recording

A reconstruction of one visitor's visit, replaying their clicks, scrolling, taps and form interactions, used to see how a page is actually used.
AnalyticstechnologyUpdated

Definition

A session recording, also called session replay, is a playback of a single visit built from the interactions a script on the page captured: scrolling, clicks and taps, pauses and activity in form fields. It is usually a reconstruction of the page and the visitor's actions rather than a screen video. The unit is one visit by one person, which separates it from analytics reports, which count events across many visits, and from the aggregate maps that the same tools often provide.

The conversion research methods guide classes recordings as a behavioural method: they show what people did on the page without asking them. They show neither the reason for the behaviour nor how common it is across all visitors, which is why recordings sit between quantitative research, which locates and sizes a problem, and attitudinal research, which explains it.

In practice: watching against one question and tallying behaviours

The method that turns recordings into evidence has a small number of fixed parts, set out in the tutorial on reviewing session recordings and heatmaps:

  • A viewing question naming a step, a device and a segment, taken from the funnel rather than a hunch.
  • Filters applied before watching, so the batch contains only visits that reached that step and left.
  • A coding sheet with one column per behaviour, such as repeated taps, back-and-forth between pages, fields cleared or re-typed, clicks on things that are not links, and long pauses before leaving.
  • An evidence rule agreed in advance for how many sessions must show a behaviour before it counts as a finding.
  • A comparison with visitors who reached the same step and continued, because a behaviour equally common among converters is part of normal use, not a cause of leaving.

The output is a counted finding, with a few example sessions attached after the count, that becomes a Hypothesis once a second source supports it. In prioritisation frameworks such as the experimentation loop, evidence of this kind is what raises a test idea's confidence score above an idea copied from a competitor.

Recordings versus heatmaps: the individual and the aggregate

A Heatmap adds up many visits into one overlay and shows where behaviour concentrates. A recording shows one visit in sequence and shows what happened before someone left. The two answer different questions and are read together: a click map might show heavy tapping on an image, and the recordings for the same segment show whether those visitors then gave up, tried another route or carried on. Neither explains motive; a poll or a usability session at the same step is how the reason is heard.

Recordings can capture what people type, and typed values are often personal data. The guide to keeping personal data out of analytics treats research tools as part of the same problem as analytics tags, and the review tutorial masks every input before any recording is collected and checks that account and order pages show no names. The recording script is also a tag that follows the site's consent decision, and it adds weight to the page it studies, so it is loaded after the page becomes interactive. This page is educational, not legal advice.

Illustrative example — not a real result

A pet insurer tallies a quote form

Copperwhim Pet Insurance sells cover for cats and dogs online, and its funnel shows that phone visitors abandon the quote form at the pet details step far more often than desktop visitors. The team writes one viewing question, filters to phone visits from the last 14 days that reached that step and left, and agrees before watching that a behaviour must appear in at least 15 of 60 sessions to count. The tally finds the pet's date-of-birth field cleared and re-typed at least twice in 27 of the 60 sessions, against 6 of 40 sessions from visitors who completed the step. Inputs were masked, so no dates were visible. The next test replaces the free-text field with separate day, month and year selectors and a format hint.

A Heatmap is the aggregate view the same tools usually offer, and the two are read for the same segment. Because recordings can capture typed values, Personal Data rules decide what the tool may collect and how it is masked. Recordings are most often reviewed on a Landing Page or a checkout step that a funnel has identified, and the counted finding they produce is the raw material of a Hypothesis. The wider set of conversion methods they belong to is mapped in the conversion rate optimisation pillar.

Common confusions

  • Watching more sessions gives better evidence. Without a question, filters and a tally, more viewing produces more anecdotes.
  • A recording shows why. It shows what the visitor did; the reason comes from asking people.
  • The memorable session is typical. Dramatic sessions are remembered and quiet ones are not, which is why behaviours are counted against a rule set in advance.
  • Recordings are the same as a heatmap. A recording is one visit in order; a heatmap is many visits summed.
  • Masking can be added later. Anything collected unmasked is already personal data that nobody needed.

More terms to look up

Where this term is used

Related