Many companies have commented that through their use of an advanced web analytics company like AT Internet, Omniture or RedEye they already have analysis of every campaign a user clicks on in their path to conversion, and as a result they struggle to see how TagMan can help them beyond that.

This post will identify the differences and why any marketer would actually need both an analytics solution and TagMan.

TagMan is an independent container tag solution.  Typically, it sits on every page within your website and will manage the serving of all your tags including campaign tags such as affiliate or PPC or email conversion tags; web analytics tags; multivariate testing tags from technologies like Optimost and retargeting tags from services like Criteo.

Via a user interface, you, the marketer can load up any of these tags passing any page parameters (such as basket values, departure dates, product IDs etc.) into the tags without needing the resource of your time-poor IT colleagues.

As part of the way TagMan works, it can track every event a user clicks or views including SEO and report that full path to conversion.  This is where I think the misconception of an overlap comes in.

The overlap misconception

The difference is that with web analytics, you retrospectively analyse the data to improve future media planning – possibly looking to attribute the credit of sales against the many campaigns a user has responded to.

With TagMan, while you still retrospectively analyse the data, you also set up an attribution model to run in real time.  On that cherished confirmation page, all the campaign tags are served conditionally through TagMan, and so depending on the campaigns the customer has responded to and the attribution model, TagMan will only serve the campaign tags which have led to the sale.

To make this seamlessly work with your CPA partners (such as performance marketing agencies or affiliate networks), TagMan goes a step further by intelligently serving the portion of campaign tags related to the portion of credit the campaign will get for each sale.

Example

By way of illustration, imagine a user clicks on a PPC link on Monday, clicks on an affiliate link on Tuesday, and clicks on an SEO link on Friday making a purchase of £90.  If you are running a flat attribution model – where the credit of the sale is split evenly by all the campaigns which drove the sale; on the confirmation page, TagMan will serve tags for all 3 campaigns, and in the case of the affiliate tag, pass a shopping cart value of 1/3 of the sale (ie £30).

In this instance, the affiliate is then able to claim their full commission on the revenue they collect through their tags and no negotiation is required after the event.

Without the tags being conditionally served, these campaign tags will be served for every sale, and the network will then claim for every sale which was generated by a click on their marketing.

Trying to develop a multiple awarding mechanism through web analytics would take analysis of each sale to calculate the correct portion of credit for each campaign, and then to present the findings to each network and partner to work out the commission payments.  All in all a fairly time consuming and messy way to work!