What should my bounce rate be in Google Analytics?

June 15, 2020 by Adrian Durow . Analytics

The disappointingly vague answer I give to this is… it depends.

Depends on the nature of your traffic. Depends on what interactions you are treating as interactive / non-bounce. Depends on loads of stuff.

Hopefully, the below post, will help answer this question for ‘pockets’ of your traffic though.

CR beats BR

I want to start by saying that conversion rates are more important than bounce rates. A non-bounce is a function of a converting session.  True, reducing bounces should lead to more conversions. Not always though.

So what should a website bounce rate be?

Firstly, try not to think about bounce rate as a whole-website metric.  Based on our experience of analysing 1000s of sites:

  • Home page traffic generally should be bouncing at under 30%, as most of it should be ‘safe’ traffic (e.g. branded search or direct)
  • Blog traffic normally bounces between 50% & 70%
  • Product page traffic. I.e. traffic which lands on those zero-moment-of-truth pages, can also bounce at a rate higher than the site average. Between 40% and 60%.

Obviously, this all depends on what these pages are supposed to do. Where the traffic comes from. What you are tracking (as interactive) on these pages. The stage of the buying journey the user is at. Some many variables.

How should you approach bounce rates then… ?

#1  Build pockets

Start by breaking down your landing traffic into pockets. Which areas of the site are users starting their journeys on?

#2  Build sub-pockets

Then break down those pockets of landings into sub-pockets – e.g. by source, by device, by user type.
Record the landing volumes and bounce rates for each pocket. For example:

#3  Calculate conversion rates of non-bouncing traffic (NBCR%)

The conversion rates of the retained traffic (the traffic which stays on the site).

E.g.
1,000 sessions land on the home page. This traffic converts at 10% (100 sessions), and bounces at 30% (300 sessions).
This means that 700 didn’t bounce.
This means that the non-bouncing traffic actually had a conversion rate of 14.3% ((100/700)*100)

Map these non-bounce-conversion-rates (NBCR) onto your sub-pockets:

#4  Pick a sub-pocket, and forecast the impact of a reduction in bounce

Pick a sub-pocket you feel is under performing, or that you feel you’ve got a good chance to reduce the bounce on.

Give that pocket a bounce rate reduction target.
Calculate how many extra non-bounces (retained sessions) you’ll get if you hit that target.
Extra non-bounces (retained sessions) x the non-bounce-conversion-rate (NBCR) = the extra conversions you’ll get by hitting that target.

Summary

Don’t think too much about your overall bounce rate.

Break down your bounce rates into pockets.  Then break down those pockets into sub-pockets.

Assess those sub-pockets.  Do they feel like they are under-performing?  Could those bounce rates by lower

Calculate the non-bounce-conversion-rates of those sub-pockets.

Pick a few sub-pockets, give them a target and a forecast conversion impact.

THEN…. then hard, interesting work starts.  HOW are you going to reduce those bounce rates???  HOW are you going to optimise the start of those journeys???

Latest articles