How to PROPERLY measure Visit Count?
Hello All,
I need to have a proper count of unique Visits. (Not unique visitors.)
I have a 30 minute window set. If I look at the Built-In Visits Dashboard, I get a count (1,858).
If I chart the Real User Visits (there are no Synthetics in our Production environment), I get another count (326).
On top of that I have a User Action Business Transaction that is split by dtCookie, which is, from what I understand, what is used to calculate Visit Count. Well, that gives me a third, different count. (1,615)
How can I properly, and accurately present the actual Visit Count?
Thanks in advance.
Best Answer
Cesar,
The Visits dashboard shows you all the visits that were in flight during the dashboard period. Some of the those visits will have started before the period, and some will finish after the dashboard period. So this isn't the right answer.
The User Action business transaction you describe is counting user actions, not visits, so that's not the right answer.
The chart (i.e., the measure) shows you the visits that have finished during each period. This is one correct answer. You might also be interested in the rate at which visits begin -- you might be able to get this using a landing page measure. This would be another correct answer.
Note that the Real User Visits business transaction (which I think is what you are charting) excludes robots as well as synthetic agents. You might try charting just the simple Visit Count measure to see what that gave you.
-- Graeme
Hello Graeme,
Thanks for your answer :)
A couple of notes:
What I'm doing with the User Action Dashboard is adding a splitting to the User Actions which is the dtCookie pulled from the Web Requests these requests originate from. The Actual User Count for this same time period is in the Millions, so I'm confident that whatever the dtCookie Count is measuring, it's definitely not User Action Count. Still, knowing that, do you still consider this a wrong approach?
The Real User Visit measure gives me the same results as the Visit Count measure, so, that number is good. (We have no Robots or Synthetics in Production).
So, it comes down to this:
If I chart the Visit Count measure and set the aggregate to Sum or Count for a 24 hour window, will this give me my actual Visits started and ended in that 24 hour period? And, should this count be the same as the dtCookie count for the same 24 period?
Thanks again!
If the browser is closed the visit ends but the cookie still exists. So if you start a new session, you would have new visit but still using the old cookie. I guess that why you are seeing the value low compared to visit count. Hope this helps.
Hello Venkat,
Thanks for the response.
That almost makes sense.... except for the fact that it's been posted over and over that dynaTrace Visits are tracked via the dtCookie. So, if dynaTrace sees a new visit, but the dtCookie remains the same, that means there is some other means that dT is using to track the proper visit count... right?
Graeme's answer tell me how to Properly measure visits, which does answer my question. But, now I'm left with the doubt of whether a visit is or isn't measured by dtCookie. Any feedback?
Welcome the
Dynatrace Community Forums
Check out the Forum User Guide and Forum Guidelines to learn how to get started.Community Member of the Month
August 2017
Announcing Dynatrace's Community Member of the Month for August 2017, Rick B.! Click here to read more! 
