Bounce Rate vs Engagement Rate
GA4 replaced bounce rate with engagement rate, then quietly re-added bounce as its mathematical inverse. Here's what changed, why your numbers moved, and which metric to actually act on.
Bounce Rate vs Engagement Rate
Engagement rate is GA4's default session-quality metric; bounce rate is now defined as 1 minus engagement rate.
In Universal Analytics, bounce rate measured single-page sessions with no interaction — a blunt signal that punished blogs and landing pages alike. GA4 replaced it with engagement rate: the share of sessions that lasted longer than 10 seconds, fired a conversion event, or viewed at least two pages. Bounce rate was initially removed, then quietly re-added in July 2022 as the mathematical inverse: 1 − engagement rate.
The two metrics are now two views of the same underlying signal. If your engagement rate is 62%, your bounce rate is 38% — by definition. The interesting question is no longer 'which is right' but 'why does it report so differently from the UA number you remember'.
The headline shift: GA4 reports a much lower bounce rate than Universal Analytics did for the same traffic. A Shopify product page that bounced at 55% in UA often bounces at 25–30% in GA4 — not because behaviour changed, but because the definition did.
Engagement rate counts a session as engaged if any one of three things happens: it lasts longer than 10 seconds, it fires a conversion event, or it generates a second page view. UA's bounce rate only required a single event after pageview. The 10-second floor catches a huge chunk of sessions UA flagged as bounces.
Typical GA4 engagement rate and bounce rate by store vertical (homepage + PDP traffic, mobile-weighted)
| Vertical | Engagement rate | Bounce rate | Avg session duration |
|---|---|---|---|
| Apparel & fashion | 58–68% | 32–42% | 1m 40s |
| Beauty & skincare | 62–72% | 28–38% | 1m 55s |
| Home & furniture | 65–74% | 26–35% | 2m 20s |
| Electronics | 55–65% | 35–45% | 1m 30s |
| Food & beverage (DTC) | 50–62% | 38–50% | 1m 15s |
| Health & supplements | 60–70% | 30–40% | 1m 50s |
Compare these to UA-era benchmarks where 40–55% bounce rates were normal on PDPs. The numbers look better in GA4 not because your store improved, but because the 10-second engaged-session threshold catches sessions that UA classified as bounces.
Why your bounce number changed overnight
Three definitional changes drove the drop. First, the 10-second engagement floor — any session that idles on a page for longer than 10 seconds is now engaged, even with zero scroll or click. Second, GA4 counts conversion events as engagement, so any session that adds to cart is engaged by default. Third, the second-pageview rule catches anyone who clicks through to a related product.
The practical effect: a returning customer who lands on your homepage, reads for 12 seconds, and leaves is now an engaged session in GA4 — but was a bounce in UA. For most Shopify stores this means GA4 bounce rates run 15–25 percentage points below what you saw in UA, and direct year-over-year comparisons between the two are misleading.
Don't compare GA4 bounce rate to UA bounce rate
If your 2022 board deck showed a 52% bounce rate and your 2024 GA4 dashboard shows 31%, you have not improved the site by 21 points. You've changed the measurement definition. Rebaseline against GA4 numbers from a clean post-migration period (we recommend October 2023 onwards, after most GA4 implementations stabilised) before you draw any trend conclusions.
Which one should you optimise for
For internal reporting, pick engagement rate. It's the metric GA4 surfaces by default, it frames the conversation positively (higher is better), and it aligns with how product teams already think about session quality. Bounce rate in GA4 carries no additional information — it's just 100 minus engagement rate.
For diagnostic work, neither metric is sensitive enough on its own. A 65% engagement rate on a product page tells you sessions hit the 10-second floor, not that visitors found what they wanted. Pair engagement rate with scroll depth, add-to-cart rate, and the related Bounce Rate in GA4 segmentation views to actually understand whether a landing page is working.
GA4 engagement rate vs bounce rate by landing-page type (apparel store, 90-day window)
Engagement rate
Bounce rate
Frequently asked questions
No. UA's bounce rate measured single-page sessions with no interaction. GA4's bounce rate is the inverse of engagement rate, which counts any session over 10 seconds, with a conversion event, or with two or more pageviews as engaged. The GA4 number runs 15–25 points lower than UA for the same traffic.
Google removed bounce rate from GA4 at launch in 2020 and re-added it in July 2022 after operator pushback. It's now defined as 1 minus engagement rate and appears in the standard reports and Explorations.
Because the 10-second engagement floor in GA4 reclassifies most sessions UA called bounces. Any visitor who idles on the page for 11 seconds is engaged in GA4 — even with no scroll or click. The drop is a definitional change, not a behavioural improvement.
Use engagement rate as your primary metric — it's the GA4 default and frames performance positively. Bounce rate adds no new information since it's just 100 minus engagement rate. Pick one and stick with it across reporting.
For most DTC verticals, 55–70% engagement rate on landing pages is normal. Beauty and home tend to run higher (60–74%), food and snacks tend to run lower (50–62%) because of shorter consideration cycles. PDPs typically sit 5–8 points below collection pages.
Not directly. Scroll is tracked as a separate Enhanced Measurement event but doesn't feed the engaged-session definition. A session can hit 90% scroll depth and still count as engaged purely because it crossed the 10-second threshold first.
Yes. In GA4 admin, go to Data Streams → your stream → Configure tag settings → Adjust session timeout, and you can change the engaged-session timer to between 10 and 60 seconds. Most operators leave it at 10 to stay comparable with industry benchmarks.
Engagement rate is a session-level percentage — what share of sessions qualified as engaged. Average engagement time is a duration — how long, on average, your page was in focus during a session. They measure different things and can move independently.
Yes. Any session that fires a conversion event — add-to-cart, begin-checkout, purchase, or whatever you've marked as a conversion — is engaged regardless of duration or pageviews. This is one reason your checkout-funnel landing pages show very high engagement rates.
Shopify's built-in analytics use different session definitions and don't expose engagement rate at all. The two systems sample, attribute, and time-out sessions differently, so a 5–15 percentage point gap on similar metrics is normal. Treat GA4 as your behavioural source of truth and Shopify as your commercial source of truth.
Get an AI expert review of your site
Paste your URL — Metricuno's AI runs the same heuristic checks a senior CRO consultant would, scoring your page and prioritising the fixes that'll move conversion fastest.