Expand my Community achievements bar.

SOLVED

Time Spent per Visits(seconds)

Avatar

Level 2

Hi.

 

According to the instruction of AA help(https://experienceleague.adobe.com/docs/analytics/components/metrics/time-spent-per-visit.html?lang=...), 

the formula for Time Spent per Visits(seconds) metric should be <Total seconds spent / (visits - bounces)>. 

 

However, I am getting some different results in my workspace. 

 

zoe_kim_0-1627279710333.png

 

61,190,025 /(309,071 - 121,858) should be 327. But, as you can see, the number of 'Time Spent per Visit (seconds)' is 332. 

 

Why does it show a different result? 

Is there any exception for this metric? 

 

Thanks.

 

 

1 Accepted Solution

Avatar

Correct answer by
Employee

The reason there is a difference is because in Time Spent per Visits(seconds) calculation, not all visits are counted - only those that have the timespent seconds on page attribute set. This is not exactly the same as visits - bounces. For many visits, there could be no timespent set, and therefore they are not included in the avg timespent calculation. This ends up with fewer of the visits being counted than just subtracting the bounces, leading to a smaller denominator and therefore a larger calculated number.

 

It should be possible to create a segment that is "visits that have timespent seconds set (great than zero)", which would allow you to get the same number to use as your divisor that we are.

View solution in original post

3 Replies

Avatar

Level 7

@zoe_kim  It is not recommended to compare the manually calculated metric value with the given in workspace as there are some other factors involved as well in the algorithm after which the value in computed in the workspace.

Avatar

Correct answer by
Employee

The reason there is a difference is because in Time Spent per Visits(seconds) calculation, not all visits are counted - only those that have the timespent seconds on page attribute set. This is not exactly the same as visits - bounces. For many visits, there could be no timespent set, and therefore they are not included in the avg timespent calculation. This ends up with fewer of the visits being counted than just subtracting the bounces, leading to a smaller denominator and therefore a larger calculated number.

 

It should be possible to create a segment that is "visits that have timespent seconds set (great than zero)", which would allow you to get the same number to use as your divisor that we are.