Nov 22009

Coupon Affiliates Can Prove Their Value

Granted, I'm not known to be a champion of affiliate marketing. However, I do acknowledge that by head-count the majority of affiliates work within the rules of the relationships, and don't intend to scam advertisers. Many affiliates drive incremental sales through hard work and smart use of paid search, organic search optimization and other legitimate activities that make them valuable partners.

I still have questions, though, about whether the sales attributed to the coupon affiliates -- which make up the vast majority of affiliate sales overall -- are actually incremental.

The data we've looked at suggests that most of these sales are not incremental, that absent the affiliate intercession the sale would have taken place on the advertiser's site without the coupon discount and without the commission to the affiliate. In other words, not only are the sales not incremental, their value to the advertiser has been significantly diminished. But we can't prove that definitively.

To me the question of incremental value hangs on the answer to one question: how did most of these customers end up on the coupon affiliate's site? If the answer is that the affiliate site has a loyal following; customers go there directly and look for good deals, then clearly the sales driven are incremental. If the customers search for a product or type of product and click on an affiliate link (paid or natural) and end up coming to your site to buy, then clearly those sales are incremental.

However, if the lion's share of the orders are from people who found the affiliate's site by typing in YOUR brand name, only a tiny fraction of the orders are incremental: those who would only buy if they found a coupon. Most advertisers prohibit affiliates from advertising on their name "Acme", but whether they're advertising or not, if the person searched for "Acme" and clicked on a natural search link for "Acme Coupons, Great Deals", that order isn't incremental. If people search for "Acme Coupon", again, I'd argue that those affiliate orders aren't incremental whether the consumers click on a sponsored link or natural link, they'd already announced their intention to buy from you.

The big coupon affiliates claim they have loyal followings and maybe they do, but I, and many retailers I know remain suspicious.

I propose a test: Coupon affiliates can easily prove their value simply by capturing the referrer string from visitors to their site (showing the last page the visitor was on prior to arriving at the affiliate's site -- including a search string if relevant) and appending it as a parameter when that traffic passes from the affiliate's site to the advertiser's. The advertisers could then see what fraction of affiliate orders come from people looking for the advertiser by name and what fraction got to the affiliate site by something other than brand search.

Mom and pop affiliates might not be able to pull this off, but the big ones certainly could. Which of them is game to accept the challenge?

Any takers?


19 Responses to "Coupon Affiliates Can Prove Their Value"
Social comments and analytics for this post... This post was mentioned on Twitter by rimmkaufman: rkgblog: Coupon Affiliates Can Prove Their Value: Do coupon affiliates drive incremental sales? I'm dou..
This already happens with existing Affiliate links just not on all Networks ... Hide Http Referrer
Leon, thanks for the info. But doesn't this software sort of imply that the referrer strings they pass might well be "spoofed"? Geesh, didn't realize the lengths to which some folks are willing to go to make a buck. Seems to me that people who are sharp enough to do this kind of thing would also be sharp enough to do valuable work for someone at a fair price and make good money with a clean conscience. Call me crazy!
As someone who spends all their time dedicated to analyzing practices in the affiliate channel, I can't help but respond to this post. I do love a discussion on policy in affiliate marketing. While I think you have an overall valid question, I feel you miss the mark overall in your conclusions. Some of this seems to be possibly related to assumptions. Assumptions can sometimes be right, but they can sometimes be wrong as well. "The data we’ve looked at suggests that most of these sales are not incremental, that absent the affiliate intercession the sale would have taken place on the advertiser’s site without the coupon discount and without the commission to the affiliate." Whatever this data source you are referencing may be, it would seem that the crux of your argument of the value of coupon affiliates rests on the suggestion that the sale would have taken place without the coupon discount. The question that leaps to my mind is HOW do you know the sale would have happened if the coupon had not been available? What type of data are you basing this statement from? What size sampling of consumer data are we talking about? What is the data demographics? I get the impression (which may be wrong) that you basing this to a large extent on how the consumer arrived at the affiliate's site. But when I'm looking at questionable affiliate practices and the value of affiliate traffic (which I do tend to think in terms of incremental sales) and assist merchants in establishing their affiliate program guidelines, I'm looking more at how the affiliate got the traffic to the merchant. That may seem like a subtle difference, but I feel it's a significant difference. Afterall, a large part of affiliate marketing is about the affilite pre-selling the merchant. You make the case that a sale is not incremental if the consumer searched for "Acme coupon", found an affiliate giving a coupon for Acme & then clicked off to the merchant to use the coupon. Further, you imply that the sale would have happened anyway if the consumer did not find an available coupon. I think it's a rather large assumption and I would be very interested to see data supporting that point of view. It probably is true in some cases. A consumer is ready to buy and they will buy anyway...they are just going to get it for less if they can. However, there will also the possibility that the consumer may have a preference of a merchant to buy from. But what is most important to them is the lowest price (hence the shopping comparison channel). If they don't find a coupon for Acme, they won't make the purchase..they will go with a competitor who does have a coupon. The marketing influence of a coupon, sale or discount is well known. There is more than one successful merchant whose entire business model is built around it. They are discounters. To me coupons or deals are marketing tools provided by the merchant to the affiliate. I fail to follow your logic that just because a consumer searches "Acme coupon" the sale is not incremental. How else would a consumer find coupons the merchant is offerring as part of their overall marketing plan to drive sales? To me it shows just the opposite. If the consumer searched "Acme coupons" then it would seem more logical that coupons are factoring into their buying decision to some degree, which would seem to factor towards an incremental sale being a play. Does value from the affiliate only come if the affiliate has a loyal following the traffic originated from? If that's the case, it would seem to negate ppc search. " sales attributed to the coupon affiliates — which make up the vast majority of affiliate sales overall" Again that would depend on the type of merchant. I've done audits for some major merchants (Fortune500) of their top 100 affiliates and they had very few coupon affiliates in that list. I do agree that merchants need to give more thoughtful considedration to how they establish and manage their coupon offers. More merchants need to be thinking about the overall marketing costs associated with the coupon campaign. There is also a need to establish acceptable practices of coupon affiliates. While the majority of coupon affiliates operate within ethical practices, as always there are some who don't. But those practices are most often related to practices once the consumer is on the affiliate coupon site. No amount of referring data to the affiliate site is going to assist the merchant with those decisions. If I equate coupon affiliates with trademark poachers, my view might be different, but I don't see them as the same at all. And in a perfect world where there aren't unscrupulous advertisers who have been known to steal affiliates business models, then maybe they should be given such potentially propriatory information as an affiliates referral traffic stats. But alas, it's not a perfect world. I am surprised at your surprise at the hidden url stuff. It's a common tactic used by folks doing trademark bidding. There are also legitimate reasons for using redirects and URL shortening. Referral spoofing is usually up to no good.
Good points, Kellie. We are missing factual data, and without it any conclusion is based merely on presuppositions. The argument that that the end user (who types in that [merchant]+[coupon] keyphrase into a search field) is already determined to buy from the merchant, and is only searching to check if any additional discounts are available, is based on a presupposition that they will buy anyway. I would be very much interested in seeing any exact data on this. It is a question that is being frequently raised, but little (if any) actual data is being referenced to support the argument. On the other hand, your assumption that there is "also the possibility that the consumer may have a preference of a merchant to buy from" and all other factors being equal, he/she may go with the merchant that does have a coupon, is also a presupposition. It seems that we can be assuming both sides of the argument at any time, but to really make a verdict (or should I better say "an educated conclusion") we do need the data (preferably from several different sources).
Kellie, Thanks so much for your detailed comment! Admittedly, the data that we have suggests, but doesn't prove the point. As a former retailer, if someone searches for "MyBrand Coupon" I'm much more interested in what drove the person to want to buy from MyBrand, than in what got them interested in a discount. If giving a discount everyone who asked for one was good marketing strategy every retailer would do that. The fact that NO retailers do that indicates it isn't smart marketing. Yet, the coupon affiliates exist and thrive. Granted the evidence is anecdotal, but I know of several retailers who halted their affiliate marketing program and found that overall sales didn't drop a penny. I know of others who paused their PPC program who not only saw their overall site sales drop, but their affiliate sales as well. Careful measurement is tough, but if I had $1,000 to spend on marketing, I'd rather spend it on activities that generate interest in my brand, not on handing out coupons to folks who already know me.
George, Your last comment seems more to the point of what I think you are trying to get at. There are definitely some merchants whose businesses can't handle widespread coupon use. They can't handle the customer acquisition cost for their profit margins. The easiest way to manage that is not offer coupons to affiliates in the first place. Secondly is to not accept coupon affiliates into their program. I do know some merchants who do that. I also know of a couple of merchants who don't allow coupon affiliates because of a branding issue..they don't want their brand assoicated with discounting. But those things are not the same as labeling coupon affiliates as trademark bidders. Now if a merchant doesn't allow coupon promotion and an affiliate does it anyway, then bad affiliate! Coupons & discounts are an integral part of some merchants business models. Coupon affiliates are a cost effective way to push out their coupons. Stores like JCP always have some coupon/sale/deal they are wanting consumers to see. People expect to pretty much always find some deal and I can see where someone might search for their deals without a full decision to make a purchase already. I've done that myself. I would agree that for a small merchant with a $1k advertising budget, coupons may not be the best bang for their buck. But it could still work. For you branding played the important decision factor, for another merchant it might be immediate sales of a particular item or reaching a certain sales volume quickly. Depends on the the goals of the advertiser and many merchant do need to give that more considered thought into making sure coupons work with overall objectives. I've seen merchants offer coupons just to increase their sign-up rate of affiliates...that's not a particularly smart business move.
Naia says:
I definitely agree with Kellie. It is really hard to judge what is buyers' intention of searching for your coupon. Besides, I believe coupon site got lots of customers that never bought from the merchant before. Hence, offering coupon can catch eyeballs and increase sales. However, it is necessary to do the research on the buyers from coupon sites to finally decide if they have incremental value.
One of the major reasons merchants object to affiliates who focus on coupons is the presence of a coupon code prompt in their checkout process. Whenever a buyer already has products in their cart and encounters that prompt many are going to search for the merchant name and coupon. As soon as that happens the coupon affiliates get easy commissions for having a page show up for that search and the merchant discounts that sale and also pays the affiliate. IMHO, this is the largest issue and I am actively seeking a solution. How can a merchant offer discounts to their mailing list without having a coupon prompt in their checkout. Removing it would eliminate the incentive to search for coupons when checking out. Discounts driven by traffic from affiliates would need to be generated in another manner. Does anyone have the solution? If you do would you please post it in the comments of the post I've linked to this comment?
Thanks for your comments folks! Geno, you're right that we don't have the data to separate what is incremental and what isn't. With display ads you can do PSA hold-out tests; with catalogs and emails hold-out tests are standard practice. But there is no good way to do a hold-out test with affiliates. Turning them all off and back on and off and on in cycles might give a sense of the real lift, but I doubt affiliates would go for it. On another post someone referenced this post and made the excellent point: the legitimate affiliates wouldn't be willing to pass the referrer information. Essentially they'd be giving away information about the legitimate means they're using to generate traffic (paid search, organic optimization, whatever) which would help their competitors. It would be great if affiliate networks actual did something besides cashing checks. Seems like they could track those kinds of figures confidentially and report on them abstracted from the individual affiliates to help clean their own house. Tough issue.
As someone who's done SEO for coupon affiliates, I can tell you that their goal is to rank for brand terms.
Eg Brand X + coupon or Brand X + discount.
I saw something about how to measure these - you look at same session visitors stopping at the coupon part of the checkout and returning in a new window or something to that effect.
Thanks for stopping by, Gab! Very interesting perspective!
Adam Riemer says:
I too have a couple of thoughts on this. I used to Manage a program on Commission Junction and we offered no coupons. We also had no leads or sales from Coupon Sites on CJ and did our best to keep toolbars and adware out. This was a multimillion dollar program that was made possible by bloggers, businesses and content Affiliates. Your comment about a large part of sales coming from coupons shows that the company needs to do a lot more recruiting for content and other types of Affiliates if they want more traffic that is non coupon. Programs can easily grow and become multi million dollar programs but they take a lot more effort and work to get there. Furthermore, I also have represented and still represent some smaller companies that are start ups and don't have any people searching for their coupons yet since not many people know about them. However, when I reached out to a couple of my top Coupon Affiliates, I asked them to give them a boost on their homepage and they were able to drive tons of sales, non coupon based, on the natural traffic they had coming in just by recommending us to their audience. Coupon Affiliates can definitely add value, especially if they have natural traffic coming in, forums and communities of people that are loyal, newsletters and other things that bring traffic to smaller start ups and new brands that are not known. Although I appreciate all points of view, Coupon Affiliates do have value and if you want other types of Affiliates as well, your Program Manager needs to do more work recruiting.


Check out what others are saying...
[...] therefore posed a challenge to Coupon Affiliates: I propose a test: Coupon affiliates can easily prove their value simply by capturing the referrer [...]
[...] Coupon Affiliates Can Prove Their Value [...]
[...] to tie this in with some more discussion and found some coverage by Scott Jangro of a challenge for coupon affiliates to prove their value.I don’t think it is the right question to ask. For me the important factors are:-Did coupon [...]
[...] the vast majority of affiliates operate above board.  Even if the value some of them contribute is debatable, they adhere to the rules and work within the system as it is.  But, there will always be a [...]

Leave A Comment