One of the indicators to measure the value of a website is the “bounce rate”.
Bounce rate is something that every web person is concerned about.
Especially if you’re tasked with improving conversions, you’ll probably check your bounce rate multiple times a day.
This time, we will explain the bounce rate in an easy-to-understand manner, including the difference from the exit rate, the guideline for the bounce rate, and how to improve it, using specific examples.
What is bounce rate?
When a user leaves after viewing only the first page (landing page, LP) they visited, this is called a “bounce.”
Bounce rate is the percentage of bounces out of total sessions.
Patterns that are counted as bounces include:
(1) Move to another site after viewing the LP
(2) Close the browser after viewing the LP
(3) After viewing the LP, do not perform any operations for 30 minutes.
(4) There are four patterns in which midnight struck while viewing an LP.
In cases (3) and (4), the user has not taken any action, so you may be wondering, “Why is this included in bounces?”
The reason why these are counted as bounces is that Google Analytics also considers patterns like (3) and (4) to be “exit.”
It’s common to find things like, “I was reading a certain page carefully and 30 minutes had passed,” or “I accessed it at 11:50 p.m., and when I was reading an LP, it was already past midnight.”
Therefore, when web managers check the bounce rate, they must also consider the content of the bounces.
Bounce rate calculation formula
You can check the bounce rate on Google Analytics, but it’s a good idea to know the calculation formula as a basic knowledge for web managers.
Let’s assume that the page you want to calculate the bounce rate for is A. The formula for calculating bounce rate is:
“Number of bounces for A” ÷ “Number of all sessions with A as LP”
It becomes.
Let’s explain with a concrete example.
Suppose you have a site X consisting of pages A, B, and C. Site X had the following sessions between May 1-5, 2019:
1st A→Withdrawal
2nd A→B→Withdrawal
3rd A→B→C→Withdrawal
4th B→A→Withdrawal
5th A→C→A→Withdrawal
If the number of sessions from May 1st to May 5th were only the five listed above, what percentage would be the bounce rate for Page A displayed in Google Analytics?
The answer is “25%”.
In order for a session to be counted in bounce rate calculations, it must meet the condition that the target page is the first page (LP).
Sessions where A is not an LP, such as the 4th day, cannot be counted in bounce rate calculations even if A’s page views are included.
In this example, there will be a total of 4 sessions in which A is the LP, days 1 to 3 and days 5.
And, as already explained, only “leaving after viewing only the LP” is counted as a bounce.
Although A is an LP on the 2nd and 3rd, it does not count as a direct return because he moved to B and C and then left.
Also, like on the 5th, when you go to LP → other page → LP → leave, it is not a direct return. In this example, only one day is counted as a bounce.
If we apply this to the formula “Number of bounces from A” ÷ “Number of all sessions with A as LP”, we get 1 ÷ 4 = 25%.
The above is how to calculate the bounce rate for each LP. Regarding the bounce rate of the entire site,
“Number of bounces across the entire site” ÷ “Number of sessions across the entire site”
Calculate with.
It’s confusing! What is the difference between bounce rate and exit rate?
Here, I will briefly explain the “exit rate”, which is a concept very similar to the bounce rate.
The percentage of all sessions on a page in which that page is the last viewed page is called the “exit rate.”
The exit patterns mentioned here are the same as the four patterns already explained in the bounce rate section.
Exit is a concept that subsumes bounce. If you leave after viewing only the LP, it is considered a bounce.
The exit rate for a certain page A is calculated as “number of exits from A” ÷ “number of all sessions including page views of A”.
Let me explain using the example used in the bounce rate calculation formula.
1st A→Withdrawal
2nd A→B→Withdrawal
3rd A→B→C→Withdrawal
4th B→A→Withdrawal
5th A→C→A→Withdrawal
“Number of departures of A” is twice on the 1st and 5th. “Number of all sessions including pageviews for A” is 5 times from 1 to 5 days. Therefore, the exit rate for page A is 2÷5=40%.
The bounce rate guideline (average value) is not very meaningful.
On the web, you can find explanations that say “the average bounce rate is around 40%,” but the number “40%” itself has no special meaning.
This is because the bounce rate varies depending on the structure of the site, the access route, the user’s search purpose, etc.
For example, suppose a woman visits a manufacturer’s website to research the ingredients of a certain cosmetic product.
If there is only one page of content for that product, the bounce rate should increase.
What if you visit Rakuten Market to check prices?
There are a large number of shops that carry the cosmetics you are looking for, and the retail prices vary. Since women are likely to go to many shops, “direct returns” rarely occur.
So, what if it was a Pera site for affiliate purposes?
Since Perasite is a “one-page complete site”, the bounce rate will be high unless you click on the link to the application form.
“High bounce rate = not necessarily bad”
Some of you may be worried that “Even though all the pages have been loaded with content with the same amount of energy, for some reason only certain pages have a low bounce rate…”
The causes of high bounce rates are not necessarily bad.
For example, if your page is very detailed about a niche topic that isn’t covered on other sites, your bounce rate is likely to be high for users who arrive at your page through organic search.
Even if users arrive from a link that provides recommended news in an email newsletter, the bounce rate will be high because they will be satisfied once they finish reading that page.
In this way, a page with a high bounce rate does not necessarily mean there is a problem with the content. Let’s calmly determine what is the cause of the high bounce rate.
How to find pages that need improved bounce rate
Pages whose bounce rate should be improved will be easier to find if you pay attention to the following two points.
(1) Relationship with length of stay
Pages with short dwell times and high bounce rates may not be providing the content most users need.
Regardless of whether it is an administrative announcement page or an index page, if you find that the page spend time is short and the bounce rate is high even though it is a content page directly related to your business, you should check the page in detail, assuming that there is some kind of problem lurking in that page.
(2) Relationship with conversion
For example, what if an e-commerce site has a bounce rate of 95%, but the CV rate of the 5% of users who do not bounce is 50%? If you improve your bounce rate, you can expect a significant increase in your CV.
However, if the CV rate of 5% who do not bounce is 1%, even if the bounce rate improves, CV may not increase much.
In this case, it is more efficient to allocate resources to improving CV rate than improving bounce rate.
3 essential points to improve bounce rate
To improve your bounce rate, pay attention to the following three points:
(1) Is it compatible with mobile first?
According to data released by Google in February 2017, 53% of users will abandon a mobile site if it takes more than 3 seconds to load.
In the mobile-first era, rich and heavy sites are avoided. The same goes for sites that don’t have a responsive design.
(2) Is there a problem with the navigation?
Inadequacies in navigation, such as broken links, under construction, incorrect internal links, and missing indexes, will also increase the bounce rate.
Navigation is directly linked to the overall usability of your site, so if you ignore any deficiencies, you may even lose potential customers.
(3) Are we providing enough content that users want?
In order to prevent users from bouncing and increase page views, length of stay, and CV, it is essential that there is enough content that users want.
In addition to arranging content that matches search keywords on the LP, we also capture the needs of users regarding the brand image of the site, saying, “I want this site to provide this much content!” It is also important to keep updating your content.
summary
◆Bounce rate is the rate at which users leave after viewing only the first page (landing page, LP) they visit.
There are four patterns that are counted as bounces: “moving to another site,” “closing the browser,” “doing nothing for 30 minutes,” and “reaching midnight while viewing an LP.”
◆The formula for calculating the bounce rate is “Number of bounces for a certain page A” ÷ “Number of all sessions with A as LP”. In order for a session to be counted in bounce rate calculations, the target page must be LP.
◆Exit rate is the percentage of all sessions on page A where A is the last page viewed. Calculated by “Number of withdrawals from A” ÷ “Number of all sessions including A’s page views”
◆The bounce rate varies depending on the structure of the site, the access route, the user’s search purpose, etc. Therefore, just because the bounce rate is high does not necessarily mean that there is a problem with the content of the page.
◆Pages whose bounce rate should be improved can be easily found by paying attention to the relationship with visit time and CV. To improve the bounce rate, there are three key points: “site loading speed,” “navigation,” and “content.”