How to track downloads in Google Analytics automatically

I have a question after reading your suggestion. Could I add your code to the front of the ga.js code and then just do the following in my html file or would this mess up the loading or would there be some other modifications to make?

< script src="http://www.mysite.com/scripts/ga.js" type="text/javascript">

Hi Bobwill,

I'm not sure I've understood your question... The idea of this code is simply to add it before the analytics code on each web page...

Implementation is really simple because you can just search through your entire site replacing your standard analytics code:

< script type="text/javascript">
var gaJsHost = (("https:" == document.location.protocol) ? "https://
ssl." : "http://www.");
document.write(unescape("< script src='" gaJsHost "google-
analytics.com/ga.js' type='text/javascript'>< /script >"));
< /script >
< script type="text/javascript">
var pageTracker = _gat._getTracker("UA-0000-1");
pageTracker._initData();
pageTracker._trackPageview();
< /script >

with:

< script type="text/javascript" src="http://www.yoursite.com/script/
gatag.js">< /script >
< script type="text/javascript">
var gaJsHost = (("https:" == document.location.protocol) ? "https://
ssl." : "http://www.");
document.write(unescape("< script src='" gaJsHost "google-
analytics.com/ga.js' type='text/javascript'>< /script >"));
< /script >
< script type="text/javascript">
var pageTracker = _gat._getTracker("UA-0000-1");
pageTracker._initData();
pageTracker._trackPageview();
< /script >

Does that clarify it?

OK, I misunderstood your explanation. The statement on your web site says "To use, place this file on all pages just above the Google Analytics tracking code." To me that meant all the "js" code (the file) needed to be inserted as the top of the "ga.js" code. Your saying to place a line of code that loads the "gatag.js" script at the beginning of the "ga.js" script. I guess I am dense. I will try to make it work.

Glad it's clear now, thanks for your comments. Hope you get it working!

Excellent Excellent Excellent Article!!!! Great Job Indeed!!!!

Hi JP!

Welcome to Good Web Practices! Thanks for posting your encouragaing message!

Seems a simple and great solution to tracking downloads. I have added the code and am just waiting for Analytics to process.

Will be using this to track downloads of PDF brochures and application forms.

Maybe I missed something but where in Analytics will this be displayed? The little graphics at the top are too small to make out.

Welcome to Good Web Practices Eagleeye!

It's great you've installed the script.

So the statistics for the downloads and external links will appear as content items. E.g. Within Google Analytics navigate to the "Content" section then select "Top content". The PDF's will be listed in here along with your websites pages. If you can't see the PDF files use the "Find URL" box and enter pdf then press "Go". This will bring up all the stats for your PDF downloads!

This script is a great way to track PDF brochure downloads, we've actually put it to use for the same purpose on other sites! Please let us know how you get on!

Remember you can also set Analytics to have the download of files set up as particular goals for your website. This can be very useful to view your success rate!

So far I have added the script to one site and it has recorded a few PDF downloads and clicks on external links. It seems to be doing the job. I will add it to our main site and see how it compares with the onclick code we are currently using.

Alex A - we are already using goals to monitor specific downloads we want to track. I don't see the above code as replacing this but more as an easier way to monitor all downloads and clicks on external links.

Delighted to learn that you have added this tracking script to one of your websites and it has worked eagleeye! Thanks for giving us the update!

hi david
i read your article, well i implemented the way you have explained it, but when i insert the onClick code in the link, my page got scattered around, i just want to track how many people are downloading my zip file.
Can you pls help me in this?
raju

Hi Raju welcome to Good Web Practices!

You don't need to add the onClick code to the link!

The only thing you need to do is add the Javascript code before your analytics code. The Javascript automatically adds the tracking for downloads so that you don't need to add it to each link. You just need to follow the instructions above... There is a discussion about someone having problems adding the code here: http://groups.google.com/group...3be1899603

Hope you'll get it working Raju!

we have built in a document repository so that true URLs and file names are obfuscated. This means that the filepath is in the db and you simply get a pointer, so that Competition_FAQ.pdf is actually a URL like http:///download/12345.

It seems that the best way to track this is to add "|download" to the line

var isDoc = path.match(/(? :doc|eps|jpg|png|svg|xls|ppt|pdf|xls|zip|txt|vsd|vxd|js |css|rar|exe|wma|mov|avi|wmv|mp3|download|forward)($|&)/);

Is that correct?

Will this be able to track and uniquely identify the downloads if they are loaded via a javascript pop up window ?

I have my site setup so that our PDF catalogs open in a new pop up window.
example url:

my sample url seems to have been blocked somehow...but it is a basic javascript pop up javascript:pic('catalogs/golight.pdf')

will it be able to track a url like this ?

Hi,
Would this code work with any stats package that uses a paste in javascript snippet to do it's counting? What would the other package have to do to mesh up with your code?

I ask because I'm testing some others in WordPress.
GA is nice but it takes way too long to find the data I want and I don't need most of what it provides.

Hi Super015, Danny and Don! Thanks very much for your posts.

With regards to the implementation of this code, you'll find the answers to the majority of your questions here: http://groups.google.com/group...3be1899603

Super015, if your download link is http:///download/12345, the download will be included in the stats as "external/http:///download/12345" This means your downloads are tracked without needing to edit the javascript.

Danny, if the downloads are loaded via a javascript window, I'm afraid that this javascript won't be able to track the downloads. You could ask ShoreTel about how to do this on the Google Analytics forums.

Don this script is designed to work with Google Analytics. I'm surprised Analytics hasn't met your needs, I'd encourage you to stick with GA as once you get familiar with the setup its fantastic.

Hi David T and Don

David - I am yet to implement it but its clearly a great piece of code,

Don, if you need your analytics almost instantaneously you may have to spend $$$ on SiteCatalyst or use something like AWStats
(http://awstats.sourceforge.net/)

The code should work with any JS-based tracking code that implements link tracking. Just substitute the call to pageTracker._trackPageview for the equivalent in the other package.

However, YGWYPF - in analytics apps that you pay for it might be unnecessary - eg SiteCatalyst allows you to track specified extensions automatically.

Brian Katz
www.vkistudios.com

Thanks David.

When I go to Google Analytics and look up content > overview or content > drilldown, I do a search for anything with "download" in the URL and they are not showing up. Do I need to remove the edit I made to the isDoc variable?

Thanks again

Hi super015 and Danny

super015, at a glance it seems like the mod to isDoc will not work because it requires that the URL must *end* with one of the extensions - note the ($|&) at the end of the regex pattern.
However, David's code, I understand, is intended where you cannot modify all the code on a site.

There must be a single piece of server-side code generating your URLs - that should be changed to generate the complete tag you need.

Danny, if your downloads are opening in another window, is that window being tracked - if it is, you will probably not want to trigger a _trackPageview because you would be counting 2 pageviews for the "download"

Brian

I added the code, but I am getting an error:

Error: uncaught exception: [Exception... "Component returned failure code: 0x804b000a [nsIDOMNSHTMLAnchorElement.hostname]" nsresult: "0x804b000a ()" location: "JS frame :: http://webauthoring.tercenim.com/gatag.js :: :: line 20" data: no]

Any ideas?

Brian, thanks for the info. I'm no longer that technical and am very rusty on regex. I do know that with our repository, every document path ends in a 5-digit number (e.g., /forward/12345 or /download/54321). How could I modify the isDoc regex for this?

super015 & Mark,

Mark, in what environment are you executing the script?

super015,
I'm an Analyst with enough JS to be dangerous

On the assumption that *all* your downloads are in the form described and that the urls end as you indicated above, it looks safe to change the Regular Expression pattern to

/(?:document|forward)/(?:[0-9]{5})(?:$|&|?)/


You want to be careful not to match URIs other than those intended, so that pattern limits matches to 5th digit either ending the URI or being followed by a query string or a &

If you want to test various scenarios, use The RegEx Coach (http://weitz.de/files/regex-coach.exe)

Brian
http://www.vkistudios.com/people.html

Hi David T and Mark

I did not look closely at Mark's error that closely but I got a similar error totally unrelated to gatag.js. It was within FireFox's code, or more likely, one of its add-ons (Web Developer)
I returned to look at Mark's error and see that unlike mine, it refers to line 20 of gata.js itself.

Mark, what are the values of all variables before line 20?

Brian
http://www.vkistudios.com/people.html

Oh! Mark - what browsers have you tested this in and in which one(s) does the error show up?

Hi David T and Mark

I've now encountered that error too:
Error: uncaught exception: [Exception... "Component returned failure
code: 0x804b000a [nsIDOMNSHTMLAnchorElement.hostname]" nsresult:
"0x804b000a ()" location: "JS frame ::
http://www.goodwebpractices.com/script/gatag.js :: :: line 20" data: no]

It occurs if the page contains anchors that have no href attribute at all.

There will have to be a test to see if hrefs has an href attribute.

Brian
http://www.vkistudios.com

PS I'm going to be at the eMetrics Summit in Toronto next week - if you're going to be there, visit me at VKI Studios' Exhibitor's booth

Last post should read "hrefs[ l ] has an href attribute". Brian http://www.vkistudios.com
-

OK I've figured out my question with a little but of regex brushup.

All of our links go to either download/ or forward/ with a 5-digit id number after the slash. Ergo the regex is:

var isDoc = path.match(/(?:download|forward)(/d{5})/);

I put this in and PDFs and downloads work great, but I can't seem to find external links. Am I missing something on where to look in top content?

Why do you make the following suggestion:
Add the following JavaScript code just before your Google Analytics tracking code:

< script src="http://www.yoursite.com/scripts/gatag.js" type="text/javascript" > < /script >

??

The order doesn't really matter from what I can tell, did I miss something?

Also, gatag.js should really be as close to the tag as possible, because if ga.js were called right after the tag, and you placed gatag.js before ga.js, then gatag.js wouldn't work at all, because document.getElementsByTagName("a") would return an empty array, even though there maybe links on the page.

I meant to say: 'gatag.js should really be as close to the [end body tag] as possible, because if ga.js were called right after the [start body tag],...'

I've tried the script on our website but I don't see any PDF's in my Stats online. Am i looking in the wrong place maybe or do I have to setup something like a goal or filters to view the stats online in Google? Thanks

Hi everyone, thanks for posting.

RobP. and Erik, you need to wait a couple of days after installing the stats to see them.

The statistics for the downloads and external links will appear as content items. To see the external pages: navigate to the "Content" section then select "Top Exit Pages". The external pages will be listed in here along with other exit pages. If you can't see the external page use the "Find URL" box and enter external then press "Go". This will bring up all the external pages the script has tracked.

Unfortunately still nothing. I have had it on our site for over 2-3weeks now. I don't know what I'm doing wrong I've checked everything I could think of. Can you check the source on any of my pages to see if it's in there correctly? The only change I made was where the gatag.js is stored which is the js folder vs. scripts. That wouldn't make any diff. would it? Site: www.uconucedd.org

sorry that should be www.uconnucedd.org

Hi Rob! Just checked out your page. You have got the link to gatag wrong! You've made a mistake in the path. Change from this:

../js/gatag.js

to this:

http://www.uconnucedd.org/js/gatag.js

That will work.

BTW I'd recommend you remove your old google tracking code and taglinks!

That was just the homepage. The others used a diff. template which I did update previously to the correct path. But I'll update the home page too. Thanks. But do you think that the hompage if incorrect would stop the other pages that have pdf's from showing up in the reports?

Thanks Again!

Hi Rob,

I'm sure we will be able to get it working on your site!

I checked the tag you added and you've done it wrong again: " < http://www.uconnucedd.org/js/gatag.js "

DOH!! :-D

I'd like to post a link on an external site that goes directly to my .zip file for download. How do I track that? It works great if you get the link from one of my existing pages.

How can I get it so that the zip file gets measured if it is downloaded directly from external sites?

Declan why don't you send them to the webpage that contains the .zip file first? Then all they have to do is click the download button and you'll be able to track it using this script.

Have you got the updated ga.js installed?

Hi Richard N - thanks for your good advice - it's always better to get people to the webpage as you can track them better... But it does happen that people hotlink files.

Hi Declan. Using Gatag.js you will be able to track on "site 2" how many people go to download the zip file on "site 1" because "site 2" will record how many people click on the link to download the zip file as the link will be included as an external link in the content stats.

If you want to see on "site 1" how many times the zip file has been downloaded from external sites, I would suggest checking the server log files to see how many times the zip was downloaded... Most hosting have AWstats installed which you could use to check this. Otherwise, take Richard N's suggestion and send the users to the html page. :-)

Wow, I installed the script yesterday and now I can actually find out which PDF files and external websites are the most popular. It certainly is worth downloading and installing.

David T, have you guys got a paypal account for donations?

Hi Richard! Delighted the script has worked as hoped! We'd appreciate any donations, here is how you can donate:
http://www.goodwebpractices.com/donate.html
Thanks again Richard.

If I use target="_blank" to open a file in a new window, I get a script error:

urchinTracker is not defined
urchinTracker(file_path);

Has anyone else had this problem?

Hi Ben! I've not had this error no... If you provide a link we could perhaps check that out for you?

http://www.esrdnetwork6.org/data-forms/test.html

David, the PDF and external tracker worked for a day and now it has stopped working.

Can you see anything wrong with the tracker on www.rockingham.wa.gov.au? The links on the right hand side should be mostly externals and PDFs are under the Employment section.

Hi Richard, try putting gatag.js just above the analytics code, so below externalsitesPDF.js. What is externalsitesPDF.js by the way?

Where in Analytics are you looking for to see how many pdfs have been downloaded?

Ben, the error occurs in IE, not in Firefox. I notice you edited the script, might it be because you changed something? Otherwise the other option would be to get ride of target="_blank"...

Thank you very much David, it works perfect for us!

I can see downloaded pdf and hits on external links. It even tracks if someone hits the back button, like: /external//history.go(-1)

We have most clicks on: /external//
What is happening in that case? Is it our "home?"

Saludos

Hi David

Thanks for that, I think I had duplicated the javascript by accident, it looks to be working now.

I was looking under Top Content and typing in pdf and external as the filter criterion.

Cheers

Richard