Skip to main content

Tracking XML, Podcasts, etc... in WebTrends

Had a hard time finding anything in the Webtrends documentation on RSS and Podcasts tracking so I just gave them a call. Here's what they had to say about the matter:

Affected Environments:

Webtrends Analytics 8.x

Question:

How can podcasts be tracked with webtrends.

Issue:

A link to a podcast is generally an XML file pointing to the location of the download.

Resolution:

By implementing SDC, and putting a call to the dcs.gif file in the podcast Description (XML file) so that the DCSURI=PODCAST:/location, the podcast will put a hit in the SDC logs when it's downloaded.

No modification to the js tag is necessary.

DCSMultitrack is not necessary.

Results will be seen in the Pages report.

EXAMPLE:

add the dcs.gif call
IMG SRC =" http://statse.webtrendslive.com/dcsttpy38dbdgj15hs4peh2gr_7t5l/dcs.gif?dcsuri=PODCAST:/WhitepapersAndGuides/&WT.ti=PODCAST:/WhitepapersAndGuides/MeasuringWeb20Technologies&dcssip=www.webtrends.com&wt.rss=aggregatorRequest&dcsext.podcastname=MeasuringWeb20Technologies"
in the podcast Description, supplementing the title and information used to describe how the podcast is identified.

Supported product versions:

WebTrends Analytics 8.x with SmartSource Data Collector

Solution:

When tracking an RSS feed, there are basically 3 events that we can track.

1. Tracking the feed itself, that is, how many times the feed went out an aggregator. We can get this by standard log file analysis from whatever server distributes the feeds. Simply edit the link that goes out to your content, (something like blah/blah2/feed.aspx?param1=foo¶m2=bar) and look for them in the logs. This may or may not be a useful statistic for you as it's going to cost a lot of page views, and it only tells you how many times the aggregators grabbed the feed from your server, not how many times it was viewed or clicked on.

2. Tracking article views. To do this you'll need a functioning SDC server. What we do here is simply insert a dcs.gif request of some type in the article description that goes out. When the article description is viewed in a tool that allows java, you can actually use the whole tag itself, in other viewers you may have to just accept using a hard coded dcs.gif request similar to the noscript portion of a standard tag (you'll loose the visitor tracking benefits of the SDC tag since we're not setting any cookies then, but you'll at least log a hit). Some viewers don't even allow images to load, so in that case we've got no way to track the article view.

3. Tracking article clickthroughs from the viewer to the article page. To do this, simply tag the target page with an SDC tag, and include whatever parameters you want in the link from the article description to the target page (for example, WT.rss=rss). You can then use whatever parameter based reporting you want in WebTrends to track this.

We can also use dcsmultitrack code to bind on-click events within the article description to track clickthroughs of different links and distinguish views from the aggregator from someone who bookmarks the link from their end browser.

Comments

Popular posts from this blog

Checking / Creating New Folders from an SMTP Import

To check if a folder exists in the CMS already you have to create a folder list. There is no direct commands to do this. dim folderExists folderExists = false set folderlist = asset.getFolderList("/path/") 'loop through folder list do while folderlist.nextEntry() if folderlist.item("_cmsLabel") = "foldername" folderExists = true exit do end if loop To create a folder in the CMS, first make a Model that only contains a folder. Then do the following: if folderExists = false then 'create folder set dict = system.createDictionary() fId = asset.create(folderName, "/Site/Global/Picture of the Week/", "/System/Models/Directory Builder", dict) end if The parameters for the asset.create("label", "Path", "Model to use", content or dictionary)

Questions about Outages

Routine CMS maintentance windows are from 6pm - 7pm, every Monday and Wednesday. Generally, the system is still available during these times, but may be unavailable for a few minutes during that period. There are other times where maintenance must be performed outside these windows due to unforeseen circumstances. We aim to provide as much notice as possible for these events, typically via email and via an alert on the login page. The routine maintenance is also mentioned on the login page on the day of maintenance.

Making references to Assets work even if the Asset has not been published yet

In the CMS, from within the current asset if you have a link to another asset and the other asset has not been published yet. The current asset will have the link disabled when published. In some cases even content will not be displayed as well. To prevent this from happening you can use this trick: asset.setPram "filter_status", "*" href = asset.getLink("path or id") Now the href will be enabled regardless of whether the asset being reffered to has been published or not.