Visits not matching Downloaded CSV Report

[rene_to]

Hi, I just came across this problem when I downloaded a daily CSV report and noticed that the Total Sum of "Visits" and "Unique Visitors" did not match. E.g I downloaded a CSV report with 50000 rows and summed up every column, everything else matches except Visits and Unique Visitors. I want to know what is causing the differences.

[brian_au · Accepted solution]

Hi Rene,

This is expected reporting behavior in Adobe Analytics in certain reports such as 'where visits or unique visitors are used as metrics, where each line item receives credit for each applicable metric. The total however is deduplicated.'

http://microsite.omniture.com/t2/kb/en_US/analytics/kb/sum-line-items-different-from-total.html

Best,

Brian

[stella_anton]

Hello,

I am posting my question here becasue I think is almost the same. However, I´ve read a lot about de-duplicated total metrics, the different types of reports and how when you try to breakdown by a certain dimension (Page, Product Name, etc) the number of Total Visits doedn´t match with the SUM of all of the components. And I get it. It makes sense. However, I don´t seem to find a logical explanation for the following case: 

                                                             

	Date	Visits
1.	Dec 18, 2016	318.159
2.	Dec 19, 2016	296.037
3.	Dec 20, 2016	284.166
4.	Dec 21, 2016	261.077
5.	Dec 22, 2016	213.449
6.	Dec 23, 2016	193.797
7.	Dec 24, 2016	127.186
	Total	1.684.490	SUM: 1.693.871

This is a standard Key Metrics report - I´ve selected "Visits" and set the date range for 1 week, 

Why is there a difference between the total that Adobe is displaying at the bottom of the table and the SUM of these numbers (in Excel). 

I mean...I´m not breaking down by a dimension, I´m not filtering, it´s just a simple trend of 1 single metrics....

Please, could someone help me understand this?

Thank you.

Stella

[EricMatisoff]

Hi Stella - the best reason I can think of would be if some visits started at the end of one day and ended at the beginning of the next day. For example:

Visit starts at 11:55pm on 12/20 and ends at 12:15am on 12/21.

Seems like that's happening in about 0.55% of your visits, which is nice and low and potentially believable. If it's worth your time, it may be interesting to break down by Minute to see what percentage of Entries are late in the 11pm hour, and even pull in Avg Time spent for those to get even crazier :)

[anniequine21]

I am having the same issue with my data. My understanding was that the credit of the visit happens when the visit ends, so why would this make the daily numbers total to a larger number than the total weekly number adobe is presenting? I am still not understanding why the weekly visit number isn't just a sum of the visits that occurred on each day of the week.