Monitoring bounce rate from Google serps

[goodroi]

I'm not a huge fan of blindly monitoring bounce rates because it can be a noisy signal BUT it can be very valuable if done properly.

1st - Don't worry about your general bounce rate. It combines too many different traffic sources and landing pages so it is too noisy for real actionable data.

2nd - Let's narrow it down to only visitors coming from Google serps because theoretically most visitors from Google should want to stay a nice long time and view multiple pages. Of course every industry/keyword/page will be different.

3rd - Now let's break it down by specific page. Remember I just said "most visitors" and I didn't say "all visitors". If you have an online calculator, a high bounce rate could be a sign of high customer satisfaction so we want to remove those false positives. If you are a content website you want to look for articles that have a high bounce rate. A high bounce rate does not guarantee a problem page but evergreen content pages with high bounce rates are much more likely to be a problem page and worth a manual review.

You want to qualify the bounce rate results. We all think our own websites are perfect but that is our ego blinding us from reality. Looking at usage data like bounce rate from Google serps helps us to spot potential trouble pages that need to be reworked to improve its value and usability.

I am not saying Google uses bounce rate but imagine if you are Google. If someone searches for an article about widgets and they click on a website A and 90% of the time they immediately return to the serps and click on website B which they don't return to the serps. If you were Google would you start thinking that website A should be demoted and website B rewarded? Even if Google is completely ignoring bounce rate, wouldn't you want to protect your brand image by trying to identify and improve your trouble pages? Protecting your brand image will make it easier to find partners for linking and cross promotions.

How do you monitor bounce rate?
What is your bounce rate from Google serps?

[netmeg]

Most of my stuff with the heavy traffic are events, and most of the people want to come in, get the deets and get out. So my bounce rates for site, page, from Google, whatever - they're always going to be on the high side. (The people who come in and look at 20+ pages are the scrapers!) It's not something I worry about for my own sites.

For some of my ecommerce clients, I mostly look at bounce rate *changes* over time, to get some idea of how we're doing on the product pages.

[mikhailblaze]

Most of my traffic comes from social media and Reddit, and I have a pretty high bounce rate. My website <snip> has content which does not cater to all users. Every time I promote a post, it targets a specific audience. After reading an entry, the audience usually don't check out other pages. I monitor it through Google Analytics.

If I were Google, I wouldn't put bounce rate into the ranking equation, though it should still be recorded in the analytics. If Google put a high emphasis on bounce rate, I don't think social media-dependent news platforms would rank that well.


My bounce rate is in the 70-90% area, but it's satisfactory to me, considering the type of website that I have.

[edited by: aakk9999 at 11:21 am (utc) on May 28, 2015]
[edit reason] Removed website name as per ToS [/edit]

[Nutterum]

The thing about the Bounce Rate is how you measure it. For example I have worked on websites with bounce-rate that is calculated on the rule : If visitor leaves before 30 seconds pass on the page its bounce. This however can be flawed way of measuring. The standard way to measure bounce, via interaction with the page, can sque your data as well. So before you start actually make decisions based on your bounce rate you need to ask the question: "What is the best way to measure the bounce rate on my website?"

[piatkow]

"What is the best way to measure the bounce rate on my website?"

You need to ask that on a page by page basis. Some may give the user the info that they need at the first glance while others may be a definite bounce if the user doesn't click through to another page.

[Broadway]

Like nutterum, I'm using "adjusted" bounce rate for this metric. I don't consider it a bounce if the visitor stays 60 sec. or longer.

[AlexB77]

IMO the only real bounce rate is when user comes to your page from SERPs, does nothing at all and in a matter of not more than 5 seconds goes back to SERPs again. Same is applicable to the referral links. Now the different story is when user comes to a page clicks download or any other button or link on the page and then returns back to SERPs, this is not a bounce rate at all. I am tracking most of the events that I need metrics for including users that stay on my pages longer than 15 seconds and my bounce rate is actually 0.27%, which is what it should be. On the other hand the time spent on the page is usually the key metrics that I would normally pay attention to in conjunction with bounce rate.

[jimsthoughts]

I tend to look at 3 types of bounces.
1. person searches google, clicks on your page, and returns to the google search with a few seconds (very bad signal, your page lost trust fast).
2. person searches google, clicks on your page, spends a lot of time on your page, and possibly even other pages of your site, but still returns to google with that same search or similar search . (still a bad signal).
I've heard the above two items called "pogo sticking"
3. person searches google, clicks on your pages, and they don't return to google..ie, they go off to Facebook, or WebmasterWorld, or they click on an external link on your site....These are the Good Signals...I believe these are now being called "long clicks".
The problem with analytics, is they don't separate out the good bounces from the bad bounces.
I tend to look at the phrases people are using to come to a page, and I ask myself, am I serving that user intent of that search, and where are people going next based on their searches?

[toidi]

How goog might use the bounce rate is really what matters the most. People come onto my site and stay there for hours at a time. This does not generate any income for goog. Is it possible that a low bounce is not a good thing?

Goog does not look at user experience the same way we do.

[honestman]

Bounce rates obviously are contingent on the type of site and pages you have and so any standard statistics are silly.

How many sites out there tease users into drilling down multiple pages, as some of even very large sites do (top 10 and top 50 lists where you have to click "next," next," interrupted by ads, etc. Clearly they work as major sites, even reputable sites, use them. But they throw off the stats, I would think. The fact that some/many people fall for such "teasers" should not penalize sites that put the info all out there on one page (or should everyone find ways to tease people to go from page to page to find the ultimate information, if any, at the end of the carrot held before them?).

I get very annoyed by such sites (and portals that use such content) but teasers have been proven to work. Why should such devious tactics be rewarded?

[IanCP]

How many sites out there tease users into drilling down multiple pages, as some of even very large sites do (top 10 and top 50 lists where you have to click "next," next," interrupted by ads, etc.

I bounce out of those sites as fast as lightening. When they appear high in the SERPs I simply overlook them thereafter as pointless "Time Wasters".

The bounce rate on my sites probably mirror my own search habits. If I search for "How to paint blue widgets green" and arrive at a page from Google which is exactly what I am looking for - one of three things happen:

1. I quickly glean the answer in under 30 seconds and return to something else on my mind.

2. I quickly glean the answer in under 30 seconds and quickly go Alt+F+P on the keyboard return to something else while the page prints out a copy. I do love print friendly pages devoid of advertisements etc. Not too many sites have print friendly pages though.

3. I quickly glean the answer in under 30 seconds and quickly again go Alt+F+P on the keyboard, return to something else while the page prints out a copy with Cute PDF which I save to a folder on the topic.

Simple! Google might consider bounce rate as something to behold, but I don't.

[honestman]

I agree with you @IanCP, but the question is whether any company that makes major income on advertising would appreciate your clever ways to extract information, and perhaps would reward sites that are "sticky" with many users who are not so wise to avoid time-wasters. After all, the web is full of "listicles" and top x lists, and other websites that take several pages for one to drill down -- and largely seem to thrive.

Are we swimming against the riptide, those of us who are not enticed by such sites -- are we being punished if we have sites that are straightforward?

[IanCP]

Are we swimming against the riptide, those of us who are not enticed by such sites -- are we being punished if we have sites that are straightforward?

Some may describe us as "Nerdy". I prefer the appellation of "Technocrat". I aim to offer my site visitors a clear shot at what they came for, advertisements on sites were a much later invention I only used to defray my costs..

Folks? Please keep the hissing and booing down to a minimum.

[toidi]

Folks? Please keep the hissing and booing down to a minimum.

I am with you. If i land on a site that has ads for what i am looking for it tells me immediately that i am on the wrong site.

If user metrics mattered, the site the ads are pointing to would show up insread of the middleman.