Apr 72009

Amazon's Gut Shot to Affiliates

Following up on yesterday's post,

There is a great deal of buzz about the impact of Amazon's decision to ban affiliates from PPC advertising.

Jeff Molander, argues that in this economy, retailers are no longer willing to accept on faith the value of affiliates. In the past the questions of whether affiliates generated incremental orders, or simply acted as a tax on orders that would have been placed anyway could be shrugged off. Today, those questions cannot be ignored.

Dick Mangina, made the case in his commentary that click arbitrage in affiliate marketing for Amazon essentially died in 2006, as competition in PPC drove up the CPCs and Amazon lowered its commissions. It's no longer even possible for affiliates to play this game.

Indeed, as we at RKG have started studying Multichannel Interactions, we find very interesting trends in the data, which point to the conclusion Amazon reached: the bulk of the orders aren't coming from PPC arbitrage on competitive keyword searches, they're coming from trademark squatting. Conversion rates of affiliate traffic are huge -- bigger than trademark search -- indicating not that the traffic is well qualified, but that the traffic is essentially already standing in line to buy from you. The coupon + the commission is simply a tax on sales that retailers shouldn't pay.

This has been our position for a long time. A well run PPC program is a more efficient use of marketing dollars than letting affiliates "do PPC for you" -- even if they don't squat on your brand. And, they do squat on your brand and have to be monitored. This monitoring is more and more complicated because the scammers have gotten quite sophisticated in targeting ads to geographies other than the corporate headquarters, doing it at night, etc.

Now, by prohibiting all PPC activities by its affiliates, Amazon can effectively squash the brand theft as well by enlisting the engines help in protecting its trademarks. If no one is allowed to use "Amazon" in their ad copy, it becomes very easy to police. Smart, very smart.

This is a gutsy move by Amazon that requires the C-level executives to really understand the channel. Too often the marketing analyst who raises these points is shouted down by the corner office folks who see nothing but the revenue tracked to affiliates. "What? Are you nuts? Affiliates are our MOST profitable marketing program!" It takes nerve to pull the plug and watch those sales migrate to organic search, brand ppc and "untracked", but migrate they will.

We'll share more of the data we've seen over the next couple of months.


19 Responses to "Amazon's Gut Shot to Affiliates"
Yup, that's what I see in the Multichannel Forensics projects I analyze. It's one big ecosystem that evolves when one channel is removed, without a corresponding drop in revenue.
Ralph Remkes says:
I understand it is not smart to let your affiliates bid on your trademark. Frankly I am surprised that Amazon even let her affiliates do this untill now. But reading all blogs and reactions I get the sense that all of a sudden PPC affiliates are the worst of the worst. Is this really so?? Can an affiliate program not utilize PPC affiliates? Amazon has a huge range of products. It seems almost impossible to set-up and maintain a good PPC campaign for all products. I have seen automated program's, but untill now the results have not been impressing. Maybe there a good program's in the USA that I don't know of(pls let me know :-) ) But to make a long story short: Isn't the middle way the right way? Do your own PPC campaigns, and what you can not manage, you let PPC affiliates take a go at it....of course with consultation on texts or brand use.....
You hypothesis George is incorrect. The sales won't simply shift to another channel unless you have done no curating within the affiliate channel. Amazon has. Amazon has built whole system like Amazon A-Store to support its affiliates. The only reasons Amazon is making this move is: -They have an internal search team and they are worried about cannibalization between the two channels -Over ten states have come out with affiliate taxes nicknamed the Amazon Tax focusing on Amazon as a target. Shifting to CPC in those states makes the most sense. Why pay search affiliates on a click when you have an internal team? Any other merchant who cuts off search as an affiliate tool won't simply have the sales rollover. Like Google, Amazon is an institution online an can dictate the way it engages with customers. Have a smaller merchant like Sephora (who actively polices their T&C search policies) try the same stunt and they would see a drastic decrease in sales.
I agree with Angel's points. Most in the affiliate community believe this is Amazon's response to state and federal legislation that is/will tax transactions that affiliates drive using PPC.
Ralph, Angel, thanks for your feedback. I totally agree that you can't cut off affiliate PPC efforts unless you can manage a robust PPC program in-house or through a top notch agency. However, if affiliate PPC efforts ARE driving material incremental sales the question becomes: why have affiliates manage your search program? A real search management firm, like ours :-), will charge in the neighborhood of 10 - 15% of ad spend under management. RKG actually places a cap on its fees, meaning the percentages are far lower for really large clients. If a company can afford to spend 25% of sales on marketing, they can roughly spend 22.5% on advertising an pay their management firm 2.5%. If the affiliate commission is 10%, they can't advertise enough to drive the top line to where it ought to be. If the retailer pays the affiliates 25% commission to incent them to push harder, the affiliate will likely only spend 18 - 20% on advertising to maximize their revenue. Again, even if affiliates have good tools to use, the incentive structure works against the retailer, contrary to popular belief. There is too much incentive for affiliates to grab only the low hanging fruit and then move on. Not to mention the amount of policing needed to keep the affiliates honest. To get maximum value out of the channel retailers should manage PPC in-house or through an agency with fees based on ad spend. Rev share deals seem safe, but they aren't.
Chris, you're wise and you kicked my butt in the NCAA Tournament Pool, so I bow to your insights on this. As I commented on the other post, Nexus isn't just an issue for Amazon; every retailer needs to pay attention to what's happening in NY. While Cuomo may not go after Mom & Pop, states are desperate for resources and brick and mortar retailers in each state are lobbying hard to create some sort of internet retail sales tax.
R.Z. says:
I got an advance notice about this change last week, and I've been told it's unrelated to the NY/CA/etc tax laws. Also, I don't think it has to do with trademark bidding. Since last summer, Amazon have already had in place an automated process to void the commissions on any sales generated through the use of the search queries that included "Amazon"/"Kindle"/etc.. That process is obviously not foolproof, but it does help minimize the problem. Hopefully for Amazon, "investing our advertising resources" means improving their PPC management tools so that they can make up for the drop in sales after May 1. Right now they have relatively poor ad copy, which leads to lower CTR, lower QS, and all the rest that goes along with that. In the long run they'll probably come out ahead, though.
Angel... good rationalization. Everyone should understand that Angel is one of the most experienced, level-headed affiliate managers around when it comes to this issue. He is also the "most documented" (allowing me to run around and publish his wisdom and experience). Example: http://budurl.com/kblp George... I guess your rationalization makes sense although at the end of the day both solutions are "outsource to the experts" and give up the ability to learn PPC for one's self. Also, your cost model (if you'll forgive me) is one that does not assure the retailer of any business outcome -- beyond hitting a marketing goal. In other words, outsourcing to a PPC agency and paying a % of budget is not only "old school" (a default payment arrangement in a world that is obviously not mass marketing oriented) but also not congruent with a profit-focused relationship. PPC agencies paid this way have little if any incentive beyond delivering traffic that converts well. In this age, conversion means next to nothing versus lifetime value of customer, return on ad spend *accross multiple channels* (understanding conversion attribution) etc. etc. You guys at RKG are experts in this so I won't lecture :) Most PPC agencies do not work closely enough with other marketing channels to facilitate such navel gazing -- which I admit is not very popular. I also argue that it will make or break many a retailer in 2009 onward. Thanks for humoring me everyone :)
Hmm. I'm sorry but can I say something again? Angel's point is fairly brilliant -- re: "affiliate curation." Nicely done, Angel. Touche!
Evan says:
I think it represents a trmandous opportunity for other retailers or Amazon's competition that do allow affiliates to use PPC Search...
Evan says:
I can't spell!
Jeff, I'm not actually wedded to a cost-based percentage either. I advocate paying your internal PPC manager or external agency a reasonable amount, however the number is derived. We reject the commission-based model for the reasons stated earlier. Paying folks fairly for their work may be "old school", but I'm pretty sure it's been around for a while for a reason. We view ourselves as part of our client's marketing team, and our incentive to perform (whether that's driving top line, bottom line, focusing on B-to-B vs B-to-C customers or whatever) stems from our desire to continue the relationship. The best employees work hard and intelligently because they love their work and are driven to excel; the notion that people are nothing more than rats responding to rewards and punishments is true of the worst among us, but not true of the folks you want to hire. I have no doubt that Angel is a master of curating his affiliate programs. The point is, only folks who are in contact with the retailer knows what their objectives are, how they view LTV, the value they place on New Customers and how those values change over time. Crowd sourcing is simply not a good mechanism for engaging in marketing activities. Moreover, why would you hire someone that you have to constantly monitor (curate)?
Marc Adelman says:
George, Wow, this topic is a hot one! It seems many of the commentators here jumped out of their seats to join this dialogue. I have only a slice of a viewpoint to interject. This dialogue can be viewed in 2 different ways and depending on which view you choose or are predetermined to choose, may give you a completely different understanding of the issue. One view is the Channel View. SEM, SEO, CSE, Affiliate, E-mail. Each Channel has a predetermined view. It is the view they are paid to see - to see the online marketing world through their channels eyes. Each channel has it's goals and has someone internally or externally assigned / paid to accomplish those goals. Therefore it is that persons duty to to drive optimal performance for their respective channel. If each channel is able to increase sales, then great right? Maybe, or maybe not. Maybe this is a flat way of viewing Online Marketing visitors & sales. The other view is the Visitor/Customer View. This view uses the Path to Purchase and LTV of the visitor/customer in aggregate to give insight into the effectiveness - true value - and potential opportunity of bottom line growth of the different online marketing channels. This view goes beyond channel into the behavioral patterns of the visitor/customer and how you as a company, through all of your channels, interact with them and their behavioral patterns. This view dares the ones looking at it to say, "Hey, if we started this thing from scratch today, knowing what we know now, how would we do it?" This question is almost impossible to ask at any company at anytime. Each channel is like a train that has built such a degree of momentum, that any major shift in that momentum may cause it to derail and lead to disaster. Additionally, there is a champion for every channel and each one will fight for it's life. (understandable in this economy! We all want to keep our jobs.) But has this economy forced us to cut a clean slice into the how we view the performance of online marketing as a whole? George, I think you nailed it by stating, "This is a gutsy move by Amazon that requires the C-level executives to really understand the channel." This view has to be mandated from the top down. I believe those who have the courage to ask the question "Hey, if we started this thing from scratch today, knowing what we know now, how would we do it?" will end up doing smarter business and provide those who don't with a model of how to survive and grow in this economy.
Greg Luecke says:
I don't follow the "trademark squatting" theory. Is the idea that the affiliates are including "Amazon" in their PPC keywords? I can't imagine that many people search for "Amazon camera" or "Amazon computer," so I don't see how this applies to a retailer like Amazon. This theory would be relevant to a manufacturer, since people would search for "Nikon camera" or "Apple computer," so these manufacturers could end up paying their affiliates for PPCs involving their own brands and trademarks. But how does this apply to Amazon?
Hi Greg, Do a search for "Amazon Coupons". Folks who do this search have already decided to buy from Amazon, but the affiliates (PPC and organic) are being credited for and commissioned for driving sales that they simply aren't driving.
Michael says:
I will note that this analysis and conclusions are incorrect as Amazon never allowed Amazon, Amazon.com or any derivative to be a part of any keyword that was purchased. You couldn't use any of their trademarks or anything that might be interpreted as a trademark in the ad text or in keyword phrases. In fact, they required all of these trademarks to be included as negative keywords to prevent inadvertent broad matching. Amazon was very diligent about finding violations to this policy and removing all revenue generated. In your example with "Amazon Coupons", Amazon would recognize and remove this affiliate within a couple days and would not pay commissions on any revenue generated. Considering that affiliates never used trademark terms to steal attribution, why do you think they made this move?
As Angel and Chris Zaharias pointed out, the prime motivation may have been the Nexus issue with NY. I wouldn't be surprised if there was also a cost/benefit analysis involved regarding the amount of time/money devoted to policing the trademark policy vs the amount of incremental benefit derived. At the time of the post any time I searched for "Amazon Coupon" I got a laundry list of affiliate PPC ads and organic links. The ads have dwindled, but the organic links are still their and I'd bet, still "driving" the bulk of their affiliate orders.


Check out what others are saying...
[...] Financial 24 - Top News placed an interesting blog post on Amazonâ [...]
[...] Following last week’s news of Amazon suddenly shutting down referral fees to affiliates resulting from PPC traffic, discussion seems to have moved to a recognition that much of the traffic about which Amazon was concerned was not resulting from PPC arbitrage on keywords but instead may have been a product of trademark squatting. [...]

Leave A Comment