There seems to be a common assumption in the affiliate world that the cookie period should be 30 days. I am concerned that not enough thought goes into determining this. What rationale is used by merchants/networks when setting cookie periods?
Sector differences
Different vertical sectors have different buying cycles and so this should influence the length of the cookie period. For example, buying a CD can be an impulse purchase, whereas people generally tend to research substantially when buying items such as holidays or loans.
My question to networks, agencies and merchants is do they take their traditional purchase cycle into account when setting the cookie period?
Rewarding affiliates fairly
Any merchant that truly buys into the affiliate model will be aware of the need to reward affiliates fairly for what they do. Whilst metrics like CPA and deduplication criteria are hotly debated by affiliates to ensure they’re getting their dues, is cookie length examined as closely?
This works the other way round too. An affiliate should only expect to be paid for a sale that they can reasonably claim to have played a major part in influencing. I struggle to see how anyone can justify that a click from an affiliate site has been the main influencer on a sale 12 months later.
Finding a balance
So how do you find common ground between these two factors. You have to weigh up the need to reward affiliates effectively for the work they have done, whilst merchants need to know they are only paying for those sales that affiliates have played the mejor part in.
So back to my point about varying this by product and understanding how consumers go about buying different items. Currently, it seems to be accepted that 30 days is the norm. Where has this come from? Is there any thought put into it?
Do networks have any data on what percentage of sales this would influence? It seems that as we speak more about moving to multi-attribution models this will become more of an issue, so time to address it now?
