The proliferation of tags

Pretty much every digital initiative a marketer undertakes involves some form of tracking to facilitate or optimise.  All ‘drive-to-web’ marketing clearly has the requirement of having campaign conversion tags or pixels sitting on the confirmation pages to track the success of the campaign, while multivariate testing and retargeting initiatives need to anonymously identify the user so they know what copy/content or creative to show.

With lots of different suppliers across different countries all requiring some form of tag on the client’s site, it isn’t surprising that so much vocabulary has been created.  I’ll try to break this down into two sections: the tag itself and the different forms it can take (including what it actually does) and then whereabouts on the page it goes.

The ‘tag’

A tag is simply a piece of code which will sit on your website. When the page is viewed by the user, the code springs to live and calls something to be retrieved from a server then the ‘tag’ will have done it’s magic.  They are often called pixels (although this is a type of tag); cookies (although this is what the tag will set on the users browser); beacons (although this is not really a tag which will sit on the page itself); universal tag (again a specific type of tag); or a container tag (like the Atlas Universal Action Tag or DoubleClick Floodlight tag).

What tags do

So, then what does this piece of code actually call?  Typically, it will do one of two things: call a 1×1 image pixel (a transparent GIF) or a JavaScript library to do something more interesting.  A 1×1 image pixel offers the most basic of telling a technology what is going on. When the page is viewed by the user, the call of the pixel can collect some parameters from the page (such as a page ID, basket value or order ID), and, when requesting the GIF from the tracking server, set a cookie on the user’s browser with an encrypted and unique identifier, and pass back to the tracking server the unique identifier (so it knows who this refers to) and any parameters or page IDs (so it knows what has gone on).

Meanwhile, cookies are simply short text files (viewed in Notepad) of encrypted information which can be read only by the technology that put them there. They don’t have any software or intelligence and cannot do anything except be recognised and be written to.  Some companies – perhaps more underhand – don’t use cookies for the fear of users deleting them, and instead they use a flash object. It works exactly the same way as a cookie, but is stored in a different place and is much, much harder to delete.

The other call a tag will make is to call a JavaScript library.  Now JavaScript is a wonderful thing in that you can write the code to do anything and therefore do anything on a page – this is perhaps one of the reasons why our esteemed IT colleagues don’t really like the idea of marketing people having this much power. However, I think that argument rather shoots itself in the foot in that an established supplier wouldn’t risk sullying their reputation by doing something which isn’t in the best interests of their client’s website.

TagMan will use a JavaScript call in a tag where we can because we (or the marketers using TagMan via the user interface) can update what the tag does without having to change the code on the website – and we all know changing code natively on a webpage can lead to huge delays.

The second issue of why IT may not be as keen on JavaScript is because a user might have disabled JavaScript to run.  Now I’m sure there is research out there with recorded stats on this, but as a straw poll, next time you are out and about with friends (not people in the industry) ask them if they know about how to turn off JavaScript.

Right, to recap. We’ve covered what the tag will call on the page, and what they’ll do on the browser. Next is where they go on the page.

Where they go

The issue of location comes down to the need for the tag to serve as early as possible in the page loading, while not getting in the way of the user experience.  Most of the time, the tag will go at the bottom of the page above the footer and still in the body of the page, although some technologies require the tag to be further up the page near the header.  If you’re confident the technology supplier of this tag will have excellent performance on their tracking servers serving the tag, you need not worry – i.e. do they serve the tags from servers in your country? Do they use a single server, or a cloud computing network? Are they sitting within a CND? If the performance is likely to be better than your own web servers, then it matters less where the tag is.  If you fear the performance might be limited, either find another supplier, or place the tag at the bottom of the page so it will be loaded once the page itself has loaded and not hinder page load times.

Tags can also be loaded within their own iFrame, essentially a parallel section which can be served simultaneously with the content of the page.  iFrames are the default technology for most container tags, including DoubleClick Floodlight and Atlas UAT, as they enable more flexibility of what the third-party tags served within the container tag can actually do (location on the page, parameters passed etc). However, they can be heavy to load. (Find out more about the impact of tags, including iFrame containers, on data accuracy and page load times in our Tag Latency Study ).

While I could go on and on about the intricacies of tags and what they do (and my family could vouch for that!), I’ll leave it there for now.  In the next blog/report, I’ll really confuse you and introduce the idea of server tags which need not go on the page at all…

TagMan’s brilliance is its ease of use. Once blank TagMan tags are installed on the site they can be configured to deploy any type of tag into the page, including those that require data to be coded in dynamically from the page. Adding, editing and removing tags can be done literally in minutes. This means no more waiting for site development cycles to come round to edit existing tags or install new ones.

Unlike the free ad server piggy-back tag solutions like Atlas UAT and DoubleClick Floodlight, TagMan can be used to deploy virtually any type of tag, including those for Web Analytics, Surveys and Multi-variant testing. Better still if these systems can share data captured from the page and by TagMan for maximum efficiency and better integration. 

Best of all TagMan can save companies thousands of wasted CPA. TagMan can track all forms of online marketing including display and natural search and can allow tags to be loaded based on the prior marketing activity. This enables companies to deploy solutions to help improve conversion for traffic responding to specific marketing campaigns as well as enabling CPA de-duplication. By only serving the tag for the company that generated the sale TagMan prevents two different CPA partners thinking they’ve generated the same sale making commission duplication a thing of the past.

TagMan makes installing new online marketing software easy. Let us enable you to take control of your online marketing performance.

The list of vendors and solutions we already support includes DoubleClick Floodlight, WebTrends, Google Analytics, Atlas action, clickdensity, TradeDoubler, Google AdWords Conversion Tracking, Yahoo Conversion Tracking, Zanox action tag, Email reaction tag, DoubleClick action, PositiveAction Tag and Metrics direct.

We’ll keep you posted about exciting new ones coming up as we continue to make TagMan a truly universal and independent tagging solution.