When dealing with time spent, there are a couple of caveats to keep in mind:
1. The last page of the visit (and subsequently visits that consist of a single page view) aren't counted in the time spent...
This is because Adobe is not a heartbeat system, the way that the time spent is calculated is basically finding the difference between the page view time stamps
Page A (12:04:38) > Page B (12:05:15) > Page C (12:07:41) > Left the Site
- The time spent on Page A is 12:05:15 - 12:04:38 (27 seconds)
- The time spent on Page B is 12:07:41 - 12:05:15 (146 seconds)
- The time spent on Page C is unknown (there is no timestamp to subtract from)
2. Given the above, this also means that the values, depending on how they round, could be slightly off.... even your own calculation could round slightly differently resulting in a slightly different value....
That said, Average Time Spent on Site (seconds) is available in workspace.... if you aren't seeing it, it's possible that your account hasn't been granted permission to that metric, or the entire metrics hasn't been added to the suite permissions....
But, the Time Spent per Visit (seconds) that you are showing in your screenshot should return the same average time spent...
I did help on a question earlier this year where the Average Time Spent on Site (seconds), I guess just due to minor changes in the calculation, was failing to work with custom date ranges... but the Time Spent per Visit (Seconds) worked just fine....
https://experienceleaguecommunities.adobe.com/t5/adobe-analytics-questions/custom-date-range-doesn-t-work-with-average-time-on-site/m-p/586632
These values should all be relatively similar, they just have minor variations in the calculation....
But if you are looking for accuracy, neither Adobe, not Google, nor any other tracking system that does a simple timestamp calculation will give really accurate values..... they should be used as a basic trend / health check... are the values going up or down; by how much; etc...
We also use Parse.ly Analytics on our sites.. this is a heartbeat system that handles time spent much more realistically... it takes into consideration when the browser or app is minimized, when the user is active in a different tab... and it sends heartbeats periodically to let the system know the user is actively engaged or not... and when the user leaves the site or closes the window/tab, it sends one final call to the system... so the time is much more accurate.
However, most companies aren't going to buy into a new system just to get more accurate time spent metrics. So my advice is to choose one primary metric, Time Spent per Visit (seconds) is more than sufficient (and if you want to use Custom Date Ranges, we know it works better than the other one), and run all your reports using it.
