What you might not know about Bounce Rate in Adobe Analytics. Calculated metrics
- Articles, Blog

What you might not know about Bounce Rate in Adobe Analytics. Calculated metrics


hi this is Andrey Osadchuk. I welcome you
to the next online training on Adobe Analytics analysis workspace. the subject
of this training is bounce rate metric. you will get to know how this metric is
defined in adobe analytics, you will get to know how differs from the bounce rate
metrics in other web analytics solutions and I will also will show you how to create
an alternative bounce rate metric that you may want to use going forward. let’s
get started. I’m an analysis workspace and I will add the page they mentioned
and they will add the out-of-the-box bounce rate metric to this table if you
thought that the bounce rate metric analytics is defined the way that when
visitors come to the site viewed only one page and then left the site and then
the bounce rate metric is incremented this is not always the case and that
will explain you why the bounce rate metric is calculated in analytics in
different way I will need to first walk you through the metrics the bounce rate
metric is based on the bounce made the bounce rate metric is defined as number
of bounces divided by number of entries bounces is a visit that consists of a
single server call entries represents the number of times the given values
captured as the first value in visit so for example if I come to the home page
and this is my first page and then I click through to the account page then
the entries metric will be incremented for the home page since this was the
first value that was captured within a visit for the page dimension if I come
to the home page and then leave the site then it could be a bounce rate so one
pound could be count but only under certain
conditions the condition is that there should be only a single server call sent
from that page so when I come to the home page once you recall is sent to
notify Divina lytx that a visitor viewed the home page until afterwards the
visitor live leaves the site then the bounces metric will be incremented
however if you have any type of home page tracking implementation that
results in an s.tl() or custom link tracking server call then the bounces
may not be incremented for example if the home page has a tracking
implementation for on click events on a certain area of the page for example
there is a forum and if a visitor clicks on the forum or in any of the fields in
the form and then the second server call s.tl() custom link server call is sent to adobe analytics then such visit will not be considered as a bounce so since the
second server call happened then the bounce will not be incremented and as a
result the bounce rate will not be considered for that visit this is the
key difference in comparison to other web analytics solutions so let me create
the bounce rate metric so that we will make sure that it’s calculated as I
explained I will call it bounce rate the bounce is divided by entries I will set
the percent for not and we’ll add those two metrics to the definition area and
we’ll save it and I will look for the new metric and
we’ll add it next to the default bounce rate metric as you can see the numbers
are the same they match each other so the formula is accurate now what if you
don’t want to consider secondary server calls as engagement so for example many
sites have Adobe Analytics implemented for tracking the scrolling events for
example when a visitors scrolls the page till 50% then a second server call was sent to notify Adobe that the visitor actually viewed the page and probably scrolled it
till half of the page then when the visitor continue scrolling and probably
scroll till the end of the page another server call is sent so that the analyst
will know how many visitors or how many visits ended with the scroll until the
very end of the page in some cases this makes sense and this makes sense from
the engagement standpoint but in some cases this doesn’t make sense as it’s
not necessarily reflect the real engagement in some scenarios you really
want the visitors either to go to another page or I probably use more than
one page after the landing page etc so in this case you will not be able to use
the bounce rate to get what you want but you can create an alternative version of
the bounce rate and for this purpose we will need to use another metric which is
called single page with it this metric is also available by default and it’s
defined as the number of visits with a single page view so here it’s important
that there was a page view doesn’t matter what type of sta
custom link server calls are happening this metric counts only pageviews server
calls so in order to update our bounce rate we’ll need to create the bounce
rate as single page Wizards divided by entries let’s do that together bounce
rates single page with its / entries again percent for month and we’ll need
to find this metric there you go and I will add it together with the bounce
rates so that we can compare the values so now we can see that for two pages
internet Wi-Fi and Internet land the original default metric of the bounce
rate and the updated custom bounce rate metrics are showing dramatically
different values and this is because on these two pages there are some on-page
events fired and this is why the bounce rate is much lower and if this is not
what you can see their engagement then you may want to use this alternative
calculated metric for bounce rate that is based on single page visits hope now
you understood the difference between the default bounce rate metric in Adobe
no ethics and the one that I showed you if you realize that you could
misinterpret your analysis in Adobe no ethics now you have a recipe that you
can use going forward let me know if you have any questions or any comments
feedbacks I would appreciate if you can add your comments under this video hope
this was helpful thank you for watching thank you for your time and talk to you
next time bye

About Ralph Robinson

Read All Posts By Ralph Robinson

Leave a Reply

Your email address will not be published. Required fields are marked *