Hit counter issues

[maestroc]

Any idea why or what circumstances might cause the Joomla article hit counts to be significantly different than what Google Analytics reports? We recently moved from a shared hosting account to a VPS and at that point the hit counts in Joomla started to be a lot less than those in GA.

Is it possible that Joomla is only counting uniques instead of all hits? It seems that all of the articles are reporting roughly 2/3 less hits in Joomla than in GA.

Thank you for any insight you might have on this.

[pe7er]

Do you use server cache?
Do you use any Joomla cache?

[JAVesey]

Is it possible that Joomla is only counting uniques instead of all hits? It seems that all of the articles are reporting roughly 2/3 less hits in Joomla than in GA.

To legitimately compare one with the other you need to make sure that the two counts are using the same metrics.

It all rather depends upon what's being counted and over what time period.

[maestroc]

It all rather depends upon what's being counted and over what time period.

Is it possible to limit what is counted in Joomla? Is there a setting for this somewhere?

[maestroc]

Do you use server cache?
Do you use any Joomla cache?

No server cache is enabled.
I turned off Joomla cache months ago just in case that was the issue but it hasn't changed anything in terms of the difference in hit counts.

[sozzled]

Of course the hit counts will be different. Google Analytics uses a small piece of Javascript to see how many hits a web-page has (including hits from bots and non browser-based web agents) and Joomla uses PHP at the time a page view (from the com_content component) to update the hit count and store the value in the database. There's nothing particularly remarkable about this.

If you look at the PHP associated with com_content you can see where and how the database information is updated. If you want more information I suggest you ask a question at GitHub and see what the J! developers have to say. I don't have the time (or, perhaps, I don't have the inclination) to look at the PHP code for the com_content component to see how this is done and, besides, however it's done, if the author of the code is still around then they can answer it quicker than me.

I accept the difference in the hit count—between what GA says and what J! says—as a "fact". I don't have any confidence in the accuracy of those counts; they're indicators of which pages are being looked at by someone or something but the numbers, themselves, can be disputed. Page caching doesn't normally affect these counters unless people enable the System - Page Cache plugin (which should never be enabled unless you have web content that is never updated) and, even then, I don't know if enabling that plugin would make any difference.