I'm not sure if I am looking at this the correct way. I need to report on how often visitors log into my site. I have a success event that fires upon successful login, along with a customer identifier. By creating a visit level segment that has: log in event exists and the customer identifier exists, I should be able to get the number of visits or instances 'rows' that this occurred.
Additionally, is it correct to assume that by dividing the visits per the above segment by the same segment at the unique visitor level would be logins per visitor? So if x = 44,000 and y = 20,000, then we can say visitors log in 2 times per month or any other time range.
Can someone confirm if the above is correct for 'how often do visitors log in'?
Not sure whether i have got your question clearly, anyhow suggesting. Correct me if i am wrong.
Since you are already capturing Customer ID, if you drop 'Login' metric against the Customer ID, will it wont give you the info that is needed?
Lets say, i login 2 times. Against my login ID, it will be 2, straight forward right?
Or if you want to understand the overall picture, then Logins / Unique Customer IDs.
Lets say, there are total 10 Unique Customer IDs in your dimension and overall login is 28. Then average would be 2.8 right?
I wont consider Unique Visitor here because they are based on Browser User Agent. If the Customer Logins via PC, Mobile and Tablet, Unique Visitor will be 3, but Customer ID will be still one in the line item.