Is referral marketing dead?
For all the marketers out there: you know that referrals has a lot of advantages — Word-of-mouth, low CPA, and more importantly, results in hyper growth for your startup, which can then show up-and-to-the-right traction to investors making you millions of dollars (or euros, if you’re on the right side of the pond), making YOU immensely rich! You then buy a huge mansion somewhere in Nice or along the Amalfi coast and..
Let’s pause right there, my friend.
Most startups and startup founders think of referrals this way (atleast till the hyper-growth part of my introduction).
“Let’s use referrals to achieve hyper-growth” is something you will hear in alleys of startups offices nowadays.
Let’s pause to question a few of hypotheses about the future based on observations:
- Referrals was a golden nugget, now it’s just another marketing channel
- Sub-channels in referrals, just like Facebook groups are for Facebook, can be found and utilized
- Getting creative especially with time and variable rewards will become a must for startups
Understanding how referrals works
I’m going to go through the workings of a basic referral mechanism in a software product. I’m going to use Dropbox as it is a well-known use case.
The above formula is the formula to calculate the number of customers at a time, t = t, from the customers at time t = 0, the viral coefficient — K, the viral cycle time — ct and the time — t.
The important elements are the viral co-efficient (K) and the viral cycle time (ct), which are the number of new users that sign up after invitation by an existing user, and the time it takes for a referral to become a user, respectively.
It doesn’t take a genius to see that ct affects the customers exponentially while K affects the customers linearly, which is why it makes sense to think of decreasing ct as a very tempting option to choose. But both these important considerations to make when trying to build a viral mechanism.
Now coming back to Dropbox, you do the following:
- Sign up on Dropbox
- Like Dropbox enough to share it with friends
- Visit the referral page and add friends’ names (an email is sent)
- Friend(s) opens the email
- Friend(s) is interested enough to click it
- Friend clicks the link and lands on the Dropbox website
- Friend signs up
- ..and this loop continues
How quickly your friends sign up directly affects the ct.
How many friends you invite and how likely they’ll sign up directly affects the K.
Referral as a user acquisition channel — Saturated, yet?
Now, referrals are nothing new. We’ve all seen them, used them, benefited from them (some immensely ;))
They mostly involve money, some other similar incentive, or the “currency” that the product uses, like storage space for Dropbox. There have been many experiments done with the referral process resulting in changing best practices. That being said, there is more room for creativity.
I honestly believe that the channel has begun to saturate because people are beginning to get bored whenever they are asked to invite friends. The only people left are those trying to mooch off of your product by spamming their friends and that is not a good way to put your product out there in front of your customers.
The next question is do these referral mechanisms need to be improved? From the standpoint of user acquisition, simply yes. Because you don’t want to be just another company offering some money for inviting 5 friends and not many seem to be giving the results that startups actually want (there’s also a consideration as to whether referral channel is an ideal channel for the startup, I’m assuming it is a yes for you).
Plainly speaking, the heyday of “traditional” referrals is over.
Need some solace? This happens to all user acquisition channels.
Source: James Currier
There are two important ideas that have been overlooked (atleast to my knowledge!) — making the referrals dynamic in nature i.e., adding a definite time-duration to the process that incentivizes users to share now and eventually reduce your ct, and adding a variable reward (I’m not talking about 10% off deals). All this while keeping the CPA (Cost per Acquisition) under control.
The referral mechanisms of tomorrow will have solved these problems intelligently.
OK, a few more assumptions before we go down this rabbit hole:
- Your product is already good and you have reached Product/Market fit
- You’re looking for Product/Channel fit
Concretely what can be done?
One way could be to find ways to decrease the viral cycle time as much as possible for your product, sector, and business type.
