Commission Splitting: Opening Pandora’s Box
The concept of splitting commission among affiliates involved in a sale or lead instead of just crediting one based on last click is coming up more and more in conversation. I understand why some advertisers think this could be a good idea. They want to make sure affiliates get rewarded for their contributions and many times last click crediting logic rewards an affiliate that didn’t necessarily convert the customer or drive incremental value. They also want to make sure that affiliates higher in the funnel are getting compensated for their efforts. The result of this commission splitting should be to drive more sales through the affiliate channel – not just split up the costs. On the surface it makes a lot of sense, but if you dig into the details, you’ll see this might not be the right solution for the problem advertisers are trying to solve.
So let’s first look at a few examples. If there is only one affiliate involved in the sale, that affiliate will earn 10%. Anytime there is more than one affiliate involved there will be a rule to split the commission up between the affiliates involved (I have yet to speak to an advertiser that wants to pay more to multiple affiliates ex. 10%+4%). So if two are involved, the last affiliate might get 6% and the second to last might get 4%. If there are three involved, the last might get 5%, second to last gets 3%, and third to last gets 2%.
Now let’s look at how this might cause problems. What happens if three affiliates are involved and the second to the last affiliate is a loyalty partner? Their business model is to split the 10% commission with their members. This loyalty partner has already told its member that they will get 5% cashback on all sales. This crediting logic breaks the loyalty business model since it only pays out 4%. Furthermore, the loyalty affiliate needs to see the total order value ($100.00) in their reporting so their loyalty member account reflects the correct order amount (more about this issue below).
In the above examples, the affiliate manager creates the split commission rules based on their subjective opinion on how things should get split up, but there are other factors that must get included in these rules. What was the actual value driven by the other involved affiliates? Can this be determined by their business model (coupon, content, loyalty, search, email) or position in the click path (first, second, third)? How far back do you look for an involved affiliate to be included in the split commission (one hour, one day, one month)? It is impossible to determine the true value of each contribution in real-time based on subjective rules. There are too many unknown factors taking place.
Still not convinced? OK, let’s look at another aspect of the click path: additional marketing channels. If we all believe that the second to the last affiliate should get something for their effort, how do we reward them when the last click was another marketing channel like paid search? Affiliate networks cannot see other marketing channels. If they do not see an affiliate as last click winner, there can be no second to the last or third to the last affiliate. In this scenario, we are unable to credit the second to the last affiliate their 4%.
I’ve got one last point to make. Accounting. Let’s say we have an order for $100. We split the 10% commission up amongst three affiliates. Do we have three new sub orders ($50, $30, and $20) with corresponding commissions ($5, $3, and $2)? Or do we have one order for $100 associated to the last click affiliate with a 5% commission and two other affiliates with commissions but no order amounts? How is this data reflected in the reporting? How do you analyze and manage partnerships based on this data?
So let’s start over. Let’s first try to determine what the issues are the advertiser is trying to solve? Are there alternative solutions that may be a) simpler, b) fairer, and c) actually solve the problem? I think the desire to give credit where credit is due is often flawed. If this were the intent, advertisers would want to pay the winner their normal rate (10%) and also pay the assists an additional percentage. So 10% to the winner and 4% to the assist = 14%. However, most advertisers only want to divide up the 10% rather than paying more.
If an advertiser has insights into the total costs and contributions across all marketing channels, they can then determine the actual value that a specific affiliate is providing. By including other key performance indicators (KPIs) like new vs. returning customers, product margin, discounts, etc., advertiser can determine how much an affiliate should be paid. If you aggregate 1-3 months of data, you can determine the average payout rate and adjust it moving forward. For example, if an affiliate was getting paid 10% on all last click sales, but after analysis, it was determined that they should get 7%, the advertiser can adjust their commission down. Conversely, if an affiliate that is typically involved in the conversion path 30% of the time but only last click 10% of the time could have their commission adjusted up to 25% with negligible cost increases to the affiliate channel. Now when they are last click, their other contributions are considered and they are compensated when they are last click.
Implementing a split-commission model is easier said than done and can end up overly complicating your affiliate program. By doing a little analysis, you can make adjustments to your compensation models and still keep your program simple for you and your affiliates. By keeping the lid on Pandora’s box, you don’t run the risk of throwing the baby out with the bathwater.
What are your thoughts on split commissions? Please provide your thoughts or experiences in the comments section below. Thanks!
Back to Resources