Cracking Invite & Earn

A determining factor for any successful business proposition is their ability to find effective ways to acquire new customers at a low cost. One such highly efficient strategy that every business hopes to crack is the referral program. We, at Faasos, have tried to fine tune this program over the last few quarters and to perfect it as best as possible to suit our requirement. Here’s the story:

Basic Formula:

Every business aims to have their customers become evangelists and invite their friends and family, who in-turn invite their friends to use the product/services, thus driving customer acquisition. Some of these will further turn evangelists and drive more users for the business.

This model includes 3 components:

  • Initial set of customers
  • The evangelists who will “invite” more people – New customer invited – say I
  • Referred people who “convert” to new users for the business – Conversions – say C

The basic principle that any Invite & Earn (I&E) program follows is increase the numberof ‘Invites’ and improve ‘Conversions’ from these invites.

 Phase I: The Basic Program:

Our first attempt at I&E program was with a semi closed wallet, which required our existing user to refer a friend to use our services. As per the I&E program, the existing user gets Rs. X when the friend transacts for the first time and the friend receives Rs. X off on their first order.

The Glitch: Since this was with a semi-closed wallet, the referred customer was required to add more money to the wallet to undertake any transaction as X was always less than what a usual transaction would be. It wasn’t a surprise that the conversions languished at ca. 20% at that time.

Phase II: The Upgraded Program:

In the second Phase, we introduced an upgraded version of the program. This version was created in manner that the rewards were directly impacting the invoice. E.g. Rs. X was directly deducted from the invoice value thus eliminating the entire ordeal of adding money to the wallet and improving customer convenience. This version definitely worked better and led to an increase of 2.5X in conversions in only a month.

The Glitch:

  • Week 1 retention – As the new users started scaling up, the first thing to do was to check the quality of the new users pouring in. The idea was to understand if they were genuine users of just some friend whose mobile was used to avail the discount / split the bill etc. The best way to determine this was to track the number of new users that were transacting again in a period of 7 days and compare it to other sources of customer acquisition such as Facebook, Google etc. (Week 1 retention). As soon as we scaled up the new users with the “frictionless” rewards program, the Week 1 retention dropped considerable indicating that the quality of new users was poor / we were unable to attract genuine ‘new users’. We needed a way to keep the rewards friction free and prevent misuse of the program to just get a discount.
  • Average Order Value – Another key feature to determine the long term success of an I&E program is the Average Order Value (AOV). Average Order Value (AOV) = Revenue / Total number of orders. Once again, this had to be similar / more than the AOV from other sources of customer acquisition (Facebook, Google etc.). It did not come as a surprise to us that the AOV for individual customer under the ‘frictionless’ program was also lower as compared to the AOV from other sources of customer acquisition.

Phase III: The Current Program:

We realized that making the process frictionless does give a disproportionate incentive to “Game” the system and order at a discount. We understood the need to pull some of it back while keeping the process clean enough to not dissuade genuine users. A few tweaks to the minimum order amounts or disallowing usage of coupons with the rewards helped us separate out the genuine users in the end.

As we can see from the Chart II below, the Week 1 retention and the AOV also increased leading to a closure of loop on the perennial question that every business will have – Do Referral Users sustain the business for a long term?

Chart I – The three versions

I&E new (1)

Chart II – The current program

I&E 2

Snapshot of the mediums used for Invite & Earn:

Untitled pie

Steady State Growth:

The current program is successfully driving good customers (Week 1 retention continues to improve along with the AOVs) and the idea was to drive higher number of new users from the same. The starting point was hence to look at the funnel and figure out where we can optimise.

Quick check on the funnel

# who visit the Invite & Earn section: 100

# who invite: 35

# of invites sent per user: 2.5  ; Hence actual invites sent: ca. 88

# who register and subsequently transact as a new user: 62

Essentially we have figured out the Invite Ratio I; ca. 88% and Conversion ration C: 62% giving us a healthy Viral Coefficient Ratio of 0.55. The best programs would have a Ratio of close to 1 (essentially every user for the business getting another user; a dream come true for every marketer J). The focus in the coming months is to scale the 0.55 as high as possible while keeping the constraints of AOV and Week 1 Retention high. Any and all of the funnel components of how many consumers visit the I&E section on the App, to driving more invites from those to maybe influencing their decision to invite more people to finally improving conversions through friendly reminders are the cards in our hand.  Watch this space to see if we can scale I&E in the coming quarters as well.

Revant Bhate,

Chief Marketing Officer

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s