They’re a magic pill. But they don’t apply to all situations.
Your product is likely to fall into one of the lines on the graph above. If you’re on the blue line the benefit to an individual user of your product increases if more users use your product.
Referral programs don’t work very well with products on the red line. For the blue line kind of products it is very hard to achieve 1 user bringing in 2 users and those 2 new guys bringing in their own 2 new guys and so on. Many companies resort to offering freebies or plain money to users to fundamentally do marketing for their company. While it is certainly not a bad idea it will not get you a hockey stick.
I often use Dropbox as an example when I speak about referral programs. Dropbox is more useful to me when I use it across multiple devices of mine. I also could find more utility if my co-workers and friends use it. Dropbox decided to structure their referral program promoting the latter use case – this is a crucial difference that must be understood.
Consider PayPal, another favourite of mine. They had two things to offer: a buyer paying the seller, and users sending money to their friends. Through all the $10 referral rewards they gave out to their users for getting their friends onboard, my guess is that they’d have done better if they have that $10 instead to add on top of money they decided to transfer to their friends. So imagine this: if you gave your friend $25 for a lunch through PayPal your friend would’ve gotten $30 and you get $5.
Mind you, hindsight is always 20/20 and do keep in mind that their belief at that time was that P2P money transfer wasn’t really that much of a business as getting merchants on eBay accept PayPal.
So while referrals are great in folklore, the magic only applies to 1% of the 1% If you’re not in that category, don’t despair because when people like a product they naturally share it socially. All you need to do is to make it easy for them. See how these guys do it.
If you’re there you’ll find that your referral program design will end up having to monetarily incentivize the users on getting other users as opposed to users finding the intrinsic value of the product increase for them.If you’re on the red line the benefit to one user doesn’t increase when other users also use the product.