Why differences arise in measuring affiliate campaigns in Google Analytics
GA is a great servant but a bad master
Google Analytics is the most widely used measuring tool not only for e-shops. The biggest advantage is that it´s free and provides all key data on website performance, traffic/visitors´ sources and behavior, and in particular a look at the economic indicators of online advertising effectiveness together from all the advertising sources and their comparison. It is great yet not perfect, failing to show a real share and benefit of individual sources in the acquisition of a given order. Or it shows but merely by its simplified view which is disputable in many respects. It is always necessary to consider if I compare pears with pears, whether my conclusions are correct, corresponding to reality. And thus whether decisions according to the GA data cannot be wrong.
Nevertheless, GA are still kind of an idol for many e-shops We often encounter arguments like “but I can only see in GA this …” or the other way round “I cannot see this anywhere in GA”. A typical example is the still widely used evaluation of an advertisement according to Last Click. Or a classical example where 1+1+1 is not equal to 3, i.e. three advertising channels, each displaying 1 conversion, but I can see in GA but 1 conversion always ascribed to just one of them. To which one, from what perspective and why can fifer and that´s what´s going on here.
Why use other tools as well
In general, the reason for differences between GA and other systems is different metrics between the respective tools. While, e.g., PPC systems use click metrics, GA use traffic. Even if many people confuse these metrics, there is a substantial difference between them. But I want to explain in this article as to why and in what GA are super and in what, in turn, treacherous or even incorrect for decision-making. This is for another article. We´ll have a look at the GA relation toward recording and evaluation of affiliate marketing performance and benefits.
How is tracking conducted in affiliate systems
Like with PPC, RTB and other advertising systems, affiliate marketing also uses special measuring and reporting tools created just for this marketing channel and its specifics. With Dognet we use Post Affiliate Pro, which has been on the market for more than 15 years and affiliate networks and directly e-shops across Europe use it.
Affiliate promotion works such that a publisher, i.e. an affiliate partner, who decides to place a form of advertising (e-shop) on its website, places in the advertising a link containing specific affiliate parameters (the publisher´s unique ID, banner ID) and an advertiser´s utm parameters just for GA. This way it has to create a link and if a potential visitor clicks it on its website, the affiliate network immediately records a click. It is written from where, to where, the exact time, the advertising format of such click. Once thus delivered visitor sends out an order, information is forwarded thanks to the sales code from the thank you page of the advertiser (e-shop) and the conversion is recorded in affiliate. It is then ascribed according to identification details (see below) to the advertiser and the publisher who brought in the customer and a commission is calculated.
How we measure conversions in Dognet
We, of course, use primarily Cookies for identification and measurement purposes in Dognet. Given the ever-growing limitations, however, we track the user´s source and record commissions based on many more parameters.
Listed here are all identifiers ordered by their use:
- A unique code of influencer coupon – a unique code generated by an advertiser for a particular publisher is atop. As a result, never copy coupon codes from other websites or social networks.
- 1st and 3rd party cookies – when clicking the publisher message (either direct or indirect), 1st and 3rd party cookies are created in the browser. Cookies are linked to a particular browser and can be simply deleted
- HTML5 Web Storage cookies – HTML5 cookies are saved in local storage and are not that easy to delete
- Flash cookies – Flash cookies are no longer specified to the browser and thus Flash cookies created in Firefox are also valid in Chrome, IExplorer, Opera, etc.
- User Agent and IP address – type and version of the customer browser in combination with IP address
- IP address – IP address of the customer only
The reason and goal is for affiliate co-operation to be as accurate and fair as possible for all the co-operating parties.
How GA records and reports performance
GA records all events in particular based on http referer using utm parameters. There are many situations, however, where GA fails to record this information and Dognet does not appear anywhere with a given order in GA, even if this was definitely the case. The reasons may be on the customer used device, the browser settings, the affiliate network traffic identification method (see above), the affiliate partner´s error, but also issues on the side of advertiser pages themselves.
Differences in GA and in Dognet? The two systems are right!
Affiliate systems use much more accurate method of measuring the affiliate channel, make use of more identification parameters, and therefore an affiliate partner is always correctly identified with a given order.
Let´s have a look at situations where you can see in Dognet in your campaign orders you “cannot find in GA”, or you can see orders but you cannot see with them anywhere on the shopping path as the Dognet source. I speak specifically of GA report: Conversions -> Multichannel filters -> Assisted conversions -> Primary dimensions=Source/medium.
When they will “not meet” in GA and Dognet = “Dognet” source will not appear with an order in GA:
- The publisher has a direct link and will not give utm – referer and Dognet as the source cannot be seen in GA
- “Someone” will remove utm parameters (proxy, antivirus, firewall) – dtto
- Re-reading the page upon approval of cookies, a wheel of fortune = visit is already kept as Direct access and thus also the conversion is written down in GA
- IP tracking – we´ll address this more in the next chapter
- The user deletes cookies between the visit and the order – GA again indicates “Direct” as the source of conversion
- Redirecting the page (301) after redesign – if the click parameters are not transferred, it will again appear in GA as Direct
- Social networks – indicated as a source is FB, YT, TW and the like
- Newsletters – as an e-mail client (Gmail, Seznam, and so on)
- Duplicity does not have canonical – Google tags itself as the source https://yoast.com/rel-canonical
- Generally known situations described in this article https://stackoverflow.com/questions/6880659/in-what-cases-will-http-referer-be-empty
Clearly, a situation may arise quite commonly when GA report does not correspond to what the affiliate system reports to you. But this is quite common with any advertising system and we´re able to clearly support at any time all background documents and parameters that identify every order and entitlement to a commission.
Briefly on IP tracking – yes or no and how to set it up
IP tracking is used to record a conversion mediated by an affiliate partner in a situation where the common conversion measurement via cookie cannot be used, for example:
- an order was completed on a device other than a click, i.e. on another device of that customer or by another family member on their device
- in deleting cookies from the browser (either intentional or unintentional) between the click and sending an order
- in absolute blocking of necessary cookies (ITP) on the browser side (Safari)
The advertiser sets it up itself as an optional campaign parameter and is entirely transparent, as is the length of Cookie. The permission and the length only depend on goals of your affiliate campaign and the knowledge of the shopping path.
For full details of IP tracking, see “IP tracking – how it works and how to set it up with Dognet”
If you as an advertiser need any reports on conversions and their evaluation, write your affiliate manager. He/she will advise you on how to find them in your account or tailor-make them for you!