Expand my Community achievements bar.

Join us for the next Community Q&A Coffee Break on Tuesday April 23, 2024 with Eric Matisoff, Principal Evangelist, Analytics & Data Science, who will join us to discuss all the big news and announcements from Summit 2024!
SOLVED

Created a segment with both new and old page name, but bounce rate and time on site had a huge difference compared with pages that are separated

Avatar

Level 1

Hi all, 

 

Not sure if anyone had the same issue as mine - recently we had an update for page names so I created a segment with both the new and old page names combined. However, under the same metrics, they are showing a huge difference for Bounce Rate and Average Time on Site.

 

New segment with both new and old page names lulao_0-1650913321441.png

New and old page names separate                           lulao_1-1650913383909.png

 

PS: segment is set up at Hit level, no filter is added

 

Please comment if you ever had the same issue and how should I fix it. Thanks a lot.

 

1 Accepted Solution

Avatar

Correct answer by
Community Advisor

It's hard to say without digging in deeper to the data... but if the names changed, and this report is showing the same date range it's a bit hard to compare... What happens if you compare a previous "full week of data" (it can be whatever day to day you want, so long as it is entirely before the name change), and an equivalent "full week of data (starting and ending on the same days of the week" and compare the two... is the data still significantly off? (Depending on when your name change occurred, you could compare more than a week... I am just making an assumption that the change was fairly recent and week might be closest cycle you can reasonable compare, but of course, a longer time frame would help even out the spikes/valleys that can impact the calculations).

 

Comparing them in the way you have shown, where the date range looks like it includes before and after the value change, means that you are getting the average bounce rate for both values for that entire time frame... it probably isn't a valid comparison because the shorter range of the two (I'm guessing the new name) is far more volatile data due to lack of history (plus all the 0s that could be included in the bounce rate before/after the change could be impacting the individual calculations when broken out this way).

View solution in original post

2 Replies

Avatar

Correct answer by
Community Advisor

It's hard to say without digging in deeper to the data... but if the names changed, and this report is showing the same date range it's a bit hard to compare... What happens if you compare a previous "full week of data" (it can be whatever day to day you want, so long as it is entirely before the name change), and an equivalent "full week of data (starting and ending on the same days of the week" and compare the two... is the data still significantly off? (Depending on when your name change occurred, you could compare more than a week... I am just making an assumption that the change was fairly recent and week might be closest cycle you can reasonable compare, but of course, a longer time frame would help even out the spikes/valleys that can impact the calculations).

 

Comparing them in the way you have shown, where the date range looks like it includes before and after the value change, means that you are getting the average bounce rate for both values for that entire time frame... it probably isn't a valid comparison because the shorter range of the two (I'm guessing the new name) is far more volatile data due to lack of history (plus all the 0s that could be included in the bounce rate before/after the change could be impacting the individual calculations when broken out this way).

Avatar

Community Advisor

Those calculations might be influenced by your Unspecified settings in the table with the separate page names. Are you using a filter or fixed elements in that latter table? Try changing the Unspecified filter and compare the results.