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

Time Spent per Session - Unique Visitors

Avatar

Level 1

Hi!

 

My team is trying to decipher the chart below re: time spent on site for unique visitors:

paigequinlan_0-1687355209776.png

A few questions:

  1. Since these are time buckets for unique visitors, if a user comes to the site and spends 3 minutes, then again later and spends 10 minutes, are they placed in both time buckets?
  2. If a user is on the site and the session "times out" - will that count for two different unique visits in different time buckets?
  3. For the average time on site for each time bucket - is that the average amount of time users spent on that site for a bucketed amount of time, moving the average as we incorporate users who spent more time on the site?

Thank you!

1 Accepted Solution

Avatar

Correct answer by
Community Advisor

Hi, let me address your questions one at a time:

 

  1. Since these are time buckets for unique visitors, if a user comes to the site and spends 3 minutes, then again later and spends 10 minutes, are they placed in both time buckets?
    • Yes, this is correct.. these buckets are based per visit. You can even see this, ID you were to add up all the UVs per bucket, you will get a much higher value than the total for the column, this is because the total de-duplicates the UVs... but for each row, the count of UVs contains any visit by a UV that fell into that time span.
  2. If a user is on the site and the session "times out" - will that count for two different unique visits in different time buckets?
    • Yes, if the user session times out, and they then come back and start loading pages, that will be a new visit with its own time spent (though in theory, if they spent 3 mins, timed out, then came back and spent another 3 mins both visits would fall under the same bucket and the user would only be counted once)
  3. For the average time on site for each time bucket - is that the average amount of time users spent on that site for a bucketed amount of time, moving the average as we incorporate users who spent more time on the site?
    • Honestly, I don't think mixing Average Time on Site and Time Spent per Visit - Bucketed is a good idea.. the numbers look sort of reasonable at the beginning... but as the buckets get larger, the Average Time on Site starts to make no sense....
    • All of Adobe's time metrics are calculated, and sometimes mixing them like this can result in some very odd results.....

View solution in original post

3 Replies

Avatar

Correct answer by
Community Advisor

Hi, let me address your questions one at a time:

 

  1. Since these are time buckets for unique visitors, if a user comes to the site and spends 3 minutes, then again later and spends 10 minutes, are they placed in both time buckets?
    • Yes, this is correct.. these buckets are based per visit. You can even see this, ID you were to add up all the UVs per bucket, you will get a much higher value than the total for the column, this is because the total de-duplicates the UVs... but for each row, the count of UVs contains any visit by a UV that fell into that time span.
  2. If a user is on the site and the session "times out" - will that count for two different unique visits in different time buckets?
    • Yes, if the user session times out, and they then come back and start loading pages, that will be a new visit with its own time spent (though in theory, if they spent 3 mins, timed out, then came back and spent another 3 mins both visits would fall under the same bucket and the user would only be counted once)
  3. For the average time on site for each time bucket - is that the average amount of time users spent on that site for a bucketed amount of time, moving the average as we incorporate users who spent more time on the site?
    • Honestly, I don't think mixing Average Time on Site and Time Spent per Visit - Bucketed is a good idea.. the numbers look sort of reasonable at the beginning... but as the buckets get larger, the Average Time on Site starts to make no sense....
    • All of Adobe's time metrics are calculated, and sometimes mixing them like this can result in some very odd results.....

Avatar

Community Advisor

You're welcome!

 

One more thing to note, Adobe's (and other Analytics systems like Google Analytics, etc) time spent is based on calculation between the time stamps of page views:

 

Page A > Page B > Page C > Exit Site

  • Time spent on Page A = Page B PV Timestamp - Page A PV Timestamp
  • Time spent on Page B = Page C PV Timestamp - Page B PV Timestamp
  • Time spent on Page C (cannot be gathered, as there is no timestamp to do math with)

 

It doesn't take into account active tab / tab in background, multi-tab browsing can mix up the timestamps (i.e. Adobe just calculates between the PV calls received... Page A on Tab 1 to Page B on Tab 2 to Page C on Tab 1, etc)

 

I usually caution people that the time spent metrics in Adobe is good for "health check"... is time spent going up, is it going down, is it holding steady when looking at trends or comparing WOW, MOM, or YOY, etc (as at volume, user behaviours shouldn't change too much)... I don't really use the calculated times as "actual" accurate time spent values...