How to Measure Scroll Depth in PostHog

PostHog already measures scroll depth for you. With autocapture on, it records the maximum scroll percentage of each pageview and attaches it to the $pageleave event as $prev_pageview_max_scroll_percentage. To analyze it, query $pageleave — not $pageview — so bounced sessions are included.

The core query: average max scroll per page

This HogQL returns the average maximum scroll depth for your top pages over 30 days:

SELECT
  properties.$pathname AS path,
  round(avg(properties.$prev_pageview_max_scroll_percentage) * 100, 1) AS avg_max_scroll_pct,
  count() AS pageleaves
FROM events
WHERE event = '$pageleave'
  AND timestamp > now() - INTERVAL 30 DAY
GROUP BY path
ORDER BY pageleaves DESC
LIMIT 20

Illustrative sample output:

path	avg_max_scroll_pct	pageleaves
/	44.8	6,210
/pricing	61.2	1,840
/blog	71.5	980

The homepage average of ~45% means anything below the halfway mark is seen by a minority of visitors. If your primary CTA is at 70% depth there, most people never reach it.

Why $pageleave, not $pageview

The maximum scroll of a pageview is only known once the visitor leaves it, so PostHog can't put it on $pageview. Querying $pageleave has a second benefit: it includes bounced single-page sessions, which is precisely the shallow-scrolling audience you most want to understand. Averaging scroll over $pageview would silently drop them.

Going deeper: a scroll distribution

An average hides the shape. This bucketed query shows how many visitors reach each depth band on the homepage — far more actionable than a single mean:

SELECT
  multiIf(
    properties.$prev_pageview_max_scroll_percentage < 0.25, '0–25%',
    properties.$prev_pageview_max_scroll_percentage < 0.50, '25–50%',
    properties.$prev_pageview_max_scroll_percentage < 0.75, '50–75%',
    '75–100%'
  ) AS depth_band,
  count() AS sessions
FROM events
WHERE event = '$pageleave' AND properties.$pathname = '/'
  AND timestamp > now() - INTERVAL 30 DAY
GROUP BY depth_band
ORDER BY depth_band

Illustrative sample output:

depth_band	sessions
0–25%	2,790
25–50%	1,560
50–75%	1,020
75–100%	840

If 45% of sessions stop in the first quarter, your above-the-fold content is doing nearly all the work — so that's where your headline and CTA have to land.

Turn the number into a decision

Scroll depth is a means, not an end. Use it to answer one question: does my primary CTA sit above or below where most visitors stop? If it's below, move it up (or add a sticky CTA) and re-run the query in two weeks to confirm more visitors now reach the action. From there, build a funnel to see what they do after they arrive.