The last five (or more) years in the adtech industry have seen an increasing tendency for major retailers to utilise attribution solutions for the purposes of understanding how effective their marketing efforts are, some going as far as deciding budgets based on the outputs of these solutions.
In theory, this makes absolute sense. Most major retailers will have countless adverts running online at any one time, some through paid search, others through display, email or on affiliate websites. Keeping track of all these channels and how they intersect is realistically something that is best done by machines purely because of the volumes of data involved. It’s no surprise that there are some brilliant providers out there; Adobe, Google 360, etc… but these solutions are rarely able to give complete answers and it’s not the fault of the software.
It’s been written before that there is:
“A flawed assumption that undermines many attribution solutions from the start: that digital media is responsible for a certain number of conversions and that the task in hand is simply to divide the credit among the various parties involved.”
Many of these solutions continue be taken as gospel though, particularly by those responsible for overseeing a retailer’s digital marketing efforts. It’s understandable that this approach might be taken considering the specific remit of a digital marketeer. I propose though, that even when limited specifically to digital advertising, the use of attribution solutions is currently hindered by an incomplete data set. The setup of every attribution solution currently in existence across the digital world does not recognise the early funnel effect of the affiliate channel. Namely, the impressions delivered by each affiliate websites is missing from the data. The table below illustrates this more clearly:
publisher tag table.png
There are currently a few reasons for this; affiliates are reluctant to share the data, retailers simply don’t know how to change the status quo given the volume of partners involved in affiliate marketing but there’s also the more basic reason that affiliates and affiliate networks have never really shouted about this being an issue. Place yourselves into the shoes of the ‘Head of Digital’ at a big-name retailer; a small percentage will have a thorough understanding of the channel but the vast majority will be using an attribution solution and factoring in the outputs of their attribution solution when deciding how much budget each channel should get. This is a problem; the effect of a channel’s influence being improperly recorded inevitably means that channel will suffer from less spend.
So what’s the solution? A basic starting point would be something like a ‘publisher tag’. A publisher tag would be a piece of JavaScript code that sits on a publisher’s site and allows a network to capture the impressions from site visitors. The networks then collate this data over the course of a day and informs whatever attribution partner an advertiser uses, about all the relevant impressions for that day:
publisher tag.png
The important difference here compared to how most networks treat impressions at the moment is the addition of factoring in impressions for text links. So much of affiliate marketing’s influence comes not from banners or widgets or even pictures of products but primarily from text links embedded within content. Only a solution similar to the one above will ever truly reveal how influential affiliate marketing really is.
The more cautious affiliates out there might instantly be worried about privacy concerns or the potential damage in letting a retailer know just how many site visitors they get. Fear not; the first iteration of a movement like this would simply tag all impressions as the ‘affiliate channel’ without reference to individual sites or placements. This would align affiliates with the practices of other channels, most notably the display channel. This is also how Facebook, Twitter and Google’s own display network currently capture and share impression based interactions.
This approach is not without its potential pitfalls though. There are questions to be answered about whether all impressions should be considered equal. For example, does a voucher code home page featuring multiple retailers offer as much value as a dedicated piece of content on a retailer’s new product line? How do either of those placements compare to a re-targeted user who sees a display ad whilst browsing their emails? These answers are best decided by experts in attribution, something I can not profess to be. The focus here is to arm the attribution solutions (and the experts) with a full data set that can help them reach informed decisions about how and where marketing budget should be spent and right now the affiliate industry does not do that.
The practical steps in making this theoretical approach a reality are surprisingly straight forward:
As an advertiser, setup the affiliate channel in your attribution solution to consider impressions (which should give you a link you can share with your network and publishers).
As a publisher, be open to the idea of putting a network tag on your site and trust that overall objective is not more scrutiny but instead levelling the playing field with other channels.
As a network, integrate with the relevant attribution API’s to ensure that captured impressions are feeding into the ecosystem that each advertiser uses.
The combined effect of publishers embracing this kind of ideology is undoubtedly more money in the industry. It will re-balance the attribution solutions used by advertisers all across e-commerce to finally understand the true value of affiliate marketing.
1 March 2019
Attribution, Tracking, Affiliate Marketing, Advertisers
COMMENTS (0)Newest First
Newest First
Preview Post Comment…