“1. Google does not have access to universal bounce rate data so are unlikely to use bounce rates. All they have is the back button stats (and GA... which they say they don't use). By their own admission, back button ain't no good metric.
2. A quality article with thousands of words is good, right? It gets a higher rank. It answers the visitor's question comprehensively. Visitors tend to close their browser after reading this page i.e. higher bounce rate. Why would Google "penalise" the page for being so great? It doesn't add up.
”
This is an argument that has been rolling around for years.
In 2013, Matt Cutts, the then public face of Google specifically said Google doesn't use a site bounce rate as a signal, and Google has said they don't use the data from our Google Analytics.
But..
There seems to be a growing argument for "back button stats". Google is ever increasingly refining search results based on user behaviour and type of queries. They have a mountain of data that can tracks user behaviour, and someone that enters your site from Googles index, then bounces back to the search index very quickly and searches or looks for another result with the same query, will arguably now carry weight. So nothing to do with GA data, just the Google God/Gorg analysing user behaviour.
The otherside of the same argument is that it's quite easy to manipulate by making a wow factor, titillation, or browse bait content to keep the user on your page regardless of his original search query. Off piste or aimless meandering being something we are all guilty of. This is personally where I am, so I'm also less convinced how or if bounce data is perceived and used.
Long and quality content is another one that seems to fuel two arguments. If it is a thorough, deep or interest article that grips a visitors attention, then are they going to "bounce away" from that page when complete, such as your closing browser argument? Or are they likely to be gripped and look for more quality content by the same author? If they book mark the page and revisit within 30 days, does that go from a negative bounce signal to a positive engagement signal?
As a straight forward directory, our bounce rate was horrendous and I used to obsess about it. But talking with other directory owners, it was clear we weren't alone. The consensus was that on a directory, particularly mobile results, a user finds the result they want, ie tel number or address and then disappear. So this is positive rather than negative.
But bounce stats are a useful to analyse in GA as they can point to problem areas in your site you might not have noticed.
The SEO bounce argument swings back and forth, which is what prompted me migrate our standalone vbulletin forum onto the directory. Since we did that, bounce rate has gone down, but traffic wise no difference. So I don't think personally it has much bearing, but the bounce debate will no doubt rumble on for many more years 