Is Splitting Cookies Going To Make Affiliates Crackers?
Posted under: The cookie attribution debate took a step forward this week with the announcement from More Niche. They’ve introduced a new system on their network that will allow a merchant to decide how to reward their affiliates. They can stick to the current model of last click wins or offer a more complex system, pretty much like that suggested recently by Vodafone.
Currently only one merchant on the network has taken the plunge into the uncharted waters of cookie attribution. Proactol, a weight loss product, will continue to offer the last referrer a commission of 30% but in addition they’re also offering a fixed CPA payment of £3.50 to the first referrer. This is described as “a thank you for introducing them to the brand”.
The basis for the decision is because “this could change the power shift away from Voucher sites (capturing brand traffic) and add more value back to the generic content affiliates”. However, a recent discussion on A4U suggested that a lot of incremental sales from a voucher site don’t come from “brand”. Perhaps More Niche/Proactol could release data to demonstrate that they are seeing sales being lost to VC sites?
The lack of data to back up the need for splitting cookies is the single biggest problem with the journey that merchants and networks are embarking upon. From an affiliate point of view I want to know that if my user clicks on my link and then makes a sale within an uninterrupted cookie period how much commission I will earn. Last click satisfies that simplicity.
However, if multi attribution is going to happen what an affiliate needs to know is how much they would earn under system 1 and how much under system 2. That needs to be demonstrated in some way, especially if, as will be the case with bigger brands, other marketing channels will be chucked into the user journey. How much money is lost from the overall affiliate pot when you start assigning influence from other areas? Who will suffer pound wise if it drops – will it be affiliates, networks, or the agency implementing the model?
The case with Proactol though is slightly easier to grasp in terms of change of earnings compared to say the currently theoretical Vodafone model. Proactol have kept last referrer commission the same at 30%, but lumped on an additional CPA for the first click – so provided you’re a first introducer then you’re going to be quids in. Nothing else changes along the food chain. They’re just paying out more – not redistributing their commission.
More Niche hope that they’re going to revolutionise the affiliate marketing industry with this technology but, is this really a revolution? Or is this just a clever way to try and attract affiliates to step forward and promote the brand? If Proactol is giving all it’s sales to VC sites then would it not just be easier not to work with them (assuming they do) and award the commission to the previous click? How long are customer journey’s who buy the product? What would the reaction have been from their top last click affiliates to say drop to 20% and 10% offered to first click?
Despite opening up more questions than answers it’s a bold move by More Niche and Proactol. If nothing else it will no doubt push along the multi attribution research and debate. What will be interesting will be to see stats in the future about what kind of impact split commissions have had on this particular program. Did it increase affiliate interest? Do affiliates feel rewarded? Do affiliates actually care? In fact that’s a good question! Do affiliates really care about split cookies or are networks/merchants the driving force behind it?
The only outcome I hope for with all this split cookie talk is that, even if it doesn’t become the norm, that we see greater transparency and greater vigilance from merchants/networks. For example a program should show who’s deduping against what, are post impression cookies in the mix, are some cookies more favoured than others etc. as well as what steps were being taken to ensure cookie stuffing and forced clicks for first, middle or last click were not a problem.
Discussion regarding this issue is ongoing over at eConsultancy where Geno Prussakov has put together an interesting analysis and poses some questions including “how do we know who really introduced the customer to the brand?”.
With the ongoing debate surrounding “the need” to replace “last click wins” I think it’s safe to predict that splitting up cookies is going to drive us all in affiliate marketing just that little bit (more) crackers.
“The lack of data to back up the need for splitting cookies is the single biggest problem with the journey that merchants and networks are embarking upon”
I think you’ve hit the nail on its head there, Jason. Data and its analysis is the only thing that should be used in order for anyone to be making decisions on value attribution. I’d go one step further and say that it’s only merchants who can do this analysis, as only they have data from across all their channels. If we looked at the affiliate channel in isolation, the attributions would be incorrect and shortsighted.
Who is to say that the first cookie drop influenced the sale more than the second, third of forth cookie drop? I guess this will produce more sites that try to get the cookie drop at all costs rather than sites that promote the brand or product with a view to getting the sale through high conversion rate. If you remove cashback and vc traffic, the last click is the most likely to have influenced the sale or final decision to buy.
Without the cashback and vc traffic, you could argue that the last cookie wins model will average itself out over time. Some times you’re the last cookie, sometimes you’re not. If you remove the last minute cookie napping by locking down the cookie drop once the products have been added to the shopping basket, most people would be happy to go for the last click wins model.
Creating a payment model that creates new problems, while partially solving existing ones isn’t a great idea.
Good thought process, and one excellent post, Jason.
Thanks for the reference too.
Agree completely with both you and Hero. The only way that answers will emerge about this is by the rigorous interrogation of data and statistical modelling to ascertain whether it is worth it for all parties.
Are any affiliates actually going to earn any more money out of it? From a merchants point of view, the reason for undertaking this would be that they felt that more “content” affiliates would sign up for their program if they had the opportunity of getting paid. Is this the case? Would it be a consideration for those affiliates that traditionally lose out on a last click basis? Would the amounts of commission being paid make it at all worthwhile?
I’m not convinced that the answer to many of these questions is yes, therefore is it worth merchants, agencies and networks investing considerable amounts of money and resource to look into this? I fear not.