While auditing an affiliate program, and going through all its major elements, we had a client ask:
Can we change our cookie length? It seems too long to me…
Their cookie length was set at 120 days.
My answer to questions like this has been the same ever since I published this blog post: since 50% of affiliate-referred visitors delete cookies by day 45, why not keep that cookie life at least at the 45 days mark (and if longer, the more affiliate-friendly you’ll appear)?
Furthermore, having analyzed the above merchant’s affiliate program’s stats, we discovered the following return days distribution (judging by the past 500 affiliate-referred sales):
So, with 92% of affiliate-generated sales happening within 2 days of the original click, does it really matter if your cookie life is set at 30, 45, or 120 days? It doesn’t for you, as a merchant; but it does for affiliates. The longer — the more affiliate-friendly the program is, and the less it looks like you’re trying to skim the commission that they could’ve earned from their referral.
Great info Geno. I had never thought of it that way or seen those figures so clearly laid out which will now give me some ammunition when attempting to get merchants to increase their cookie.
I do realise that cookie length has always been relatively cosmetic and makes a program attractive, but in terms of affiliate recruitment it is a great tool to, as you said “Appear more affiliate friendly”.
In saying that do you think that an extended cookie of 365 days is actually appealing to affiliates? In my experience whenever I saw this cookie length I always felt that maybe the merchant was a bit “desperate” or it had no sway on my decision as I understood cookies and their limits and how networks record the information and that timeframe would essentially have been useless.
Zane, thank you for your comment, and glad you’ve found this post of use. Yes, I do believe that, all other variables equal, 365 or lifetime cookie life affiliate programs are more affiliate-friendly, and are perceived that way by affiliates.
Great post Geno. I have been looking at this type of data for many years and it is very common for the vast majority of conversions to occur with the first 48 hours. I also agree with the argument that if most of the conversions happen within 48 hours, why not set a referral period of 180 days to make the campaign as attractive to partners as possible.
However, if you look at it from an advertiser’s perspective, they sometimes prefer to synch their marketing channels to the same referral period to get better ‘apples to apples’ reporting across all of their channels. This is why many larger brands have shorter referral periods (session based to 7 days) – they are tracking all of their media with the same referral period. Since conversions outside the first 48 hours are so few (5% in your example) it doesn’t always make sense to increase the referral periods across all of their media channels.
Most advertisers see the ‘conversion influence window’ (yes I just made this term up) as less than 48 hours. In other words, the consumer is responding to an ad and buying because they saw that ad within the conversion influence window. It is hard to argue that a consumer was influenced by an ad they saw 30 days ago and that is why they are now converting.
I just thought I would chime in with a different perspective on this issue.
Todd, thank you for chiming in. Your input is always greatly valued and appreciated.
You’re making some great points (which do help put things into perspective).
However, when advocating longer cookie life I keep those affiliates in mind who may introduce a new prospect to a merchant, yet the prospect won’t turn into a customer until, say, Halloween (if it’s a Halloween-related merchant) or right before Christmas (if we’re talking gifts-related merchants), or when there is an appropriate occasion for those flowers or gift basket to be sent out… Yes, the majority of sales that we register fall inside that “conversion influence window” (I love the newly-coined term, by the way) of the first 48 hours. And yes, in some instances (e.g.: with lead gen programs, or products that aren’t seasonal, or those that don’t take long to decide) shorter cookie lives may be acceptable. However, since merchants really have little to lose (with people deleting cookies, and often converting into customers within the first couple of days), I would still stick to my conclusions.
Once again, thanks for your input, Todd. Always good to see you commenting in my blog.
Privet
I have read your blog now for some time and it actually is very informative – many thanks – as an Oxford graduate I can always appreciate the competition 🙂
Very interesting article on Cookie lifetime – but probably not a viable proposition when the affiliate is targeting repeat customers.. much as I’d love this.
How about an article on affiliate transparency? I often wonder whether passed traffic equates to conversions appropriately or some vendors are somewhat less than honest – always a worry.
keep up the good work!
Spasiba bolshoi!
PS. Ya govoroo pa-ruskki nyet ochin horosho – rabota moskva 2 goda 🙂
Ed, very good point on “repeat customers”. We do have to keep in mind, however, that many will still compare your proposition to your competitor’s before committing to buy again…
An article on affiliate transparency? That topic (as well as overall transparency in affiliate marketing — not only on the affiliate side, but also from merchants, affiliate networks, agencies that manage affiliate programs) can easily add up into a research/white paper or even a book. 🙂 Maybe I will tackle it from one side, or another, in one of my upcoming posts. Thanks for the idea.
PS: What college in Oxford were you at? I studied languages and philosophy at Oxford between 2000 and 2002.
I am a private SEO, only do my own sites and do affiliate marketing and also own merchant stores. So I am on both sides all of the time.
All of my affiliate programs have lifetime cookies and IP tracking. As an affiliate I do not even entertain an an affiliate opportunity unless they have lifetime or at least 180 day cookies.
This works best when you sell a product that needs service items. You know you are likely to get return visitors/customers. An example would be a specialty coffee maker with their own pod cups to brew 1 cup. You get the commission on the main product and residuals on coffee and other maintenance parts.
If you can setup like this, you will draw hordes of affiliates that will brand your product for you in a relatively short period of time. it is worth having the lifetime commissions over the years.
Short cookie life periods are OK for merchants selling quick purchase items. They are little short of a scam when used by merchants selling products for which the average consumer is likely to take a while before deciding what to purchase. The best example is travel products, where people may take months between looking round for the holiday they may take next year, then actually booking it. So all those travel blogs and info sites that help consumers make a choice during the early stages of decision making get done out of commission on a large proportion of sales which they actually initiated. For the travel sector, cookies of anything less than 90 days just serve to let merchants take orders in the knowledge that the affiliate driving the custom to their site will not be entitled to commission on a lot of sales, because the sale is “outside the cookie life”. A great idea for the merchant, but a pretty poor idea in terms of affiliate-friendliness.
But arguments about the optimal cookie life period are actually a sideshow. The real question is “are cookies necessary at all”, and the answer to that question is a clear NO.
One of the most successful affiliate programs in the travel sector, one that has helped propel the merchant in question to the no. 1 spot in its niche, is the Booking.com affiliate program. And it does not use cookies at all – at least not the kind of cookie that determines how long affiliates can get credited for. Booking.com use, in place of affiliate cookies, affiliate tracking numbers which never expire. So as long as a customer clicks on to Booking.com from an affiliate’s link, then makes a purchase, that affiliate will get the commission – even if the affiliate link came from an email or a page bookmarked months or years previously.
That’s honest dealing