Visits and entries nearly match when analysing a segment

[florianr7064714]

Hi,

we have identified a problem with entries measured within a segment in AdHoc Analysis.

We wanted to measure the visits and entries for an URL. The URL has different pagenames. That is why we combined them in a segment. The result was that the number of Visits and Entries was nearly the same, which is not logical.

We then tried to measure Visits and Entries for just one of the pagenames and the number of Visits and Entries were different. Then we applied a segment with just that pagename and it also showed the same result as before - Visits and Entries were nearly the same.

This Screenshot shows visits and entries measured by the pagename.

This screenshot shows visits and entries measured by the segment that contains the same pagename as above.

Here you can see the segment.

As you can tell, the numbers are different in the first screenshot but the same in the second one.

Can someone help us to solve that problem?

Thanks a lot!

[Urs_Boller]

what is your breakdown? what are you looking at as dimension rows? are there the URL's?

[florianr7064714]

The breakdown is "Day". In the first screenshot the dimension is just the pagename. In the second screenshot the dimension is a segement that contains just that pagename. You can see the segment in the third screenshot.

It also makes no different if we display "Hits" or "Visits" within the segment.

[Urs_Boller · Accepted solution]

sorry to not answer earlier, the question is not that easy but I think I know what happens here.

basically, "entries" does not belong to a certain page. instead, it is a global counter which can be used just by any dimension you want. you would use "page" the most (eg. for a bounce rate calculation), but you can use it for anything else like "language", "login status" or whatever you capture in your dimensions.

now back to your tables: whenever you apply "entries" to a table, it looks what "dimension" you are interested.

filter 1: dimension item

in the first case there are "day" and a certain "page". the number there indicates, how many times the page was the first captured "page" on a given "day" throughout all visits. having a look at the second row you see that you had 37'992 visits on that page and 26'994 "entries" (visits where the "day" / "page" combination was the first that has been captured).

I hope everything is clear until know because now it gets tricky ...

filter 2: hit segment

you changed the way your "entries" counter works in the second table. applying the hit segment with the desired page doesn't have any effect on the "visits", since you are still looking at the same visits (people who visited the page). but now, the "entries" counter has only one "dimension" left to show a number, which is "day". that means as soon as you have the page dimension in the segment, it is not visible for the entries counter anymore. at the end, the counter shows a number, how many times a certain "day" was the entry date of a visit. basically this should be as high as visits, the small difference is due to some midnight breaks... however, the second table the entries does not count for the same as in the first...

I know, this is really tricky, but I hope that I could explain it. let me know if anything is unclear.