Today we are asking about preferred payment structures and also whether it is a good move for networks to offer different payment structures to their merchants.
Do you think a network-wide approach to commission attribution is actually beneficial to merchants?
Tina Judic (Artemis8): It can only be a good thing if a network is able to facilitate commission attribution. I don’t necessarily believe all networks need to manage attribution in the same way – each network is looking for their USP and this could be one network’s defining offering. However, there must be a relative consensus across the networks regarding the communication and education of multi-attribution models. An advertiser must be sure they are able to truly manage attribution, and they fully understand the potential impact of rolling out attribution to their affiliate base.
Helen Southgate (Sky): No. I think this is risky as it would vary by sector and by merchant. I think any multi-attribution commission structure would need to be well thought out, understood and bespoke to each client. Also, there is the massive issue of additional cost, it just won’t work if it costs more, hence perhaps why there has been no major movement towards this just yet.
Patrick Lynch (dgm): I think network wide approach is the first step, but we should be looking at an online industry wide model. Looking at attribution across Affiliate, Search, Email and Media Buying activity as well so that the spend is going to the areas which are generating the most value to the merchants.
Peter Rowe (affilinet): I think it’s debatable and really depends on how the attribution (multi and/or value) model manifests itself. There are so many different scenarios, affiliates and advertisers it’s very difficult to predict exactly how this will play out. Does simply splitting up the commission between 3 affiliates instead of 1 result in a “better” customer – not necessarily. I believe there will be as many casualties as winners if we move to this payment mechanism. Currently, with the exception of the excellent research by Vodafone, I don’t feel there is sufficient data to make an informed decision on whether or not it will be beneficial.
Hero Grigoraki (Webgains): Networks should come together and produce best practice guidelines when it comes to applying value attribution models within the affiliate channel, which to use to educate merchants. For instance, we would expect the merchants to look at all their online activity, rather than isolate the affiliate partners. This is a project that the IAB is currently undertaking and Webgains is participating in this.
Kevin Edwards (Affiliate Window / IAB Chair): All merchants are different and value metrics will vary. I think we’re in a transitional stage where networks are being challenged to demonstrate the value of the affiliate channel more than ever before. This will manifest itself in different ways. There is no one size fits all approach but we will all have to be more flexible, open and innovative about what information we share and how we do it to make better, more informed decisions.
James Little (AffiliateFuture): Perhaps. I feel that some merchants have started to go too far in de-duping and attribution however. Lots of merchants are considering de-duping against SEO for Christ sake, and I’m pretty sick of it. Affiliate Marketing, because of it’s nature, is the only place that you can take money back from the marketing activity – you can’t get money back form google or display activity and the industry seems to be content on trying to destroy itself by letting merchants get away with some awful de-duping against search. Changing the last click wins model may, to be honest, make it worse.
Jason Dale (Loquax): I think a network needs to offer a merchant alternatives to last click and explain the positives and negatives with offering an alternative payment structure.
What concerns me is that there's not a buzz from affiliates to change models - and I do wonder why?
Julia Stent (Top10.com): Definitely not. I've talked this through with a lot of merchants across the market and if there's one thing we're all agreed on, it's that there is no cookie-cutter solution. Whereas multi-attribution might work for Vodafone, it would be nothing like as good a fit for eBay as their Quality Click Pricing, and likewise neither of these might be suitable for other merchants like Red Letter Days or Lastminute.com. It's up to the individual merchant to look in detail at their own programme and judge what will work best for them.
I believe the role of the network is to educate merchants about the options available to them, provide expertise to guide them to find the best option for them and give them tools to implement their chosen model. A lot of the networks are doing very well with this - for example I know Affiliate Window and TradeDoubler have done extensive research into customer journeys via affiliates. However, to the best of my knowledge MoreNiche are the only network to openly provide functionality for merchants to easily implement their choice of a multi-attribution scheme, which I think is fantastic.
I hope this is a catalyst for the major players to follow suit and make this the default technology offering for an affiliate programme.
Duncan Popham (Total Search Solutions): I'd like to refer again to multi-channel attribution as, frankly, without this being included, there is little benefit to anyone. I think maybe the more appropriate question is maybe, do I think that merchants having a sensible attribution policy is actually beneficial to networks...!
Finally, your preferred payment model and why.
Maria O'Flynn (TradeDoubler): I think this is more relevant to a publisher or advertiser, but from our perspective, a payment model that stimulates greater volume (such as multi-click attribution where it induces more ‘first’ clicks that result in transactions) but also that retains efficiency of administration, and that old chestnut – transparency.
Tina Judic (Artemis8): The last click model still works well for Artemis8 – it’s straightforward and ensures an immediate understanding of your actions. However, multi-attribution can certainly work in the right environment. For example, in the paid search arena, many affiliates are allowed to bid on brand – because it’s high volume and high conversion – as long as there is an investment into generic terms. This is not dissimilar to the multi-attribution approach. Instead of illustrating investment, with multi-attribution we would be able to identify, and apportion value to, clicks that were delivered earlier in the purchase path. However, if this is siloed to affiliate marketing, this is where the model falls down – as there is no correlation to other search marketing activity, let alone the broader digital mix.
Helen Southgate (Sky): I would like to see a move to a mult-attribution model but across online marketing channels, not within affiliate. I think this would be fair. How to do this without increasing costs however is a whole new discussion!
Patrick Lynch (dgm): We don't believe in a preferred payment model as this should be done on a campaign by campaign basis. Merchants in conjunction with their network should be able to build a campaign that meets their business needs. Whether that is rewarding the first and last click, or giving a higher percentage to specific affiliate channels. Having this flexibility will allow merchants to reward volume drivers but also increase their footprint in channels that can influence the purchase.
Clarke Duncan (PaidOnResults): Currently happy with last click wins, but will welcome new ideas. As you know Paid On Results have operated additional payment models in tandem with last click win and we have always had a multi-commission reward system in place but as yet no Merchant has wanted it, indeed we have our own ideas on how we can make an even better system but we will be keeping them under wraps for the time being.
Peter Rowe (affilinet): At this stage I will stick with last click. Why? As Kevin @ AWIN stated in last week's article it’s the “least worst” mechanism. When dealing with a large number of partners I feel it’s important to have simplicity and transparency. Whilst last click is not perfect it is significantly better than attribution models. Moreover, I feel many advertisers are already running “value” based commission structures rewarding on lifetime value, contract length and repeat purchasing as a few examples.
Hero Grigoraki (Webgains): We believe there is far bigger benefit to correct program setups, with profitability in mind and evaluation of consumers referred by each affiliate, rather than try to cut down on affiliate spend by applying a complicated formula that no one quite understands. With this in mind, we believe the last referrer model will continue prevailing, at least for the foreseeable future, but with more tight control and better evaluation processes. CPA, and CPE in a short while, are accountable and straightforward – which is exactly what the affiliate channel should continue being.
Kevin Edwards (Affiliate Window / IAB Chair): Value attribution, intelligent commissions setting or whatever you want to call it are all logical steps in online evolution. I firmly believe multi-attribution is the wrong answer to part of the question. Too many affiliate campaigns are badly run or only partially understood by advertisers sotTo move to an untried multi-attribution model that won’t necessarily address concerns would be a mistake. Ultimately we need to fully understand where we are at the moment, as well as dispelling a number of myths about cookie overwriting along the way. If we don’t understand the data how can we be expected to make the right decisions?
James Little (AffiliateFuture): _ My opinion is that the last click model works, and that we won’t get multi network buy-in, etc, so whilst split commission might work for a very small / niche network, it thankfully won’t happen across the board._
With that said, I do however think that the networks that innovate are the ones that will reap the most awards – so those that offer reporting to merchants on the affiliates involved in the journey will have a strong product as clever merchants will be able to set their CPAs around this data, getting more of an understanding of how certain affiliates are involved in the journey.
Jason Dale (Loquax): I think that's an impossible question because there's no data to compare last click versus an alternative. If it was shown that an alternative model boosted our commissions, rewarded us for driving traffic, and for increasing databases of various merchants - then that'd be great.
However, I think last click tends to work globally (some you win, some you don't). It has it's problems, but probably a lot less than any alternative.
The question I'd ask is why is there the clamour to change things? What is leading this charge and for what?
Is it because agencies/merchants are paying out more on the affiliate channel and want to sort out deduping issues? Is it because affiliates are concerned due to losing out to VC sites?
What I mean is, let's assume it's the latter, - is there actually a simpler solution than split commissions? e.g. remove click to reveal, encourage all affiliates to promote voucher codes, get merchants to sort their sites out etc.
Julia Stent (Top10.com): As previously explained, there is no one-size-fits-all payment model for the affiliate channel. Merchants need to make informed decisions about what will incentivise their affiliate partners to drive incremental value for them. As long as it is appropriate for their situation, my preferred payment model would be multi-attribution. I put forward one example of how this could work dubbed 'Payment on Influence' in a white paper I released about the research I did at Vodafone.
The basic idea is to reward up to five affiliates per transaction with with commissions weighted according to their stage of the journey, incentivising the first and last interactions most heavily. Multi-attribution frees affiliates from constantly 'chasing the last click' and rewards them for activity at any stage in a journey.
If introduced in the right sectors, I think multi-attribution would vastly increase affiliate earning potential and stimulate innovation while still tying in the key attraction to advertisers of affiliates being all about getting paid for the sales you influence.