Revamping Referral Program

Ads can buy attention, but it's only the users who create momentum of product growth. Here's how we designed referrals to become the engine of sustainable product-led growth.

CONTEXT

Probo as a prediction market is loved by 52M traders and sees the highest influx of traffic during the IPL Cricketing Season. For the IPL 2025, the team had planned more scale, for the first time, TV ads that would give use 40X impressions as compared to the Facebook or Google ads.

However, more the marketing budget, more is the pressure to maximise the visitor-to-trader conversion on Probo.

To solve for high intent users, we explored Referral as an important channel:

It has a lower CAC (customer acquisition cost)
Higher quality of user acquired because of the word of mouth promotion

PROBLEM

The old Referral program at Probo was not working efficiently primarily because the reward was not substantial enough for the referring user.

The average amount per referral was less a ₹1 (~$0.6 to be exact), our Referrals by WAU (weekly active users) stood at just an abysmal ~5%

There were missed opportunities - like absence of gratification, no way for Referring user to track their rewards which meant that the motivation was low.

To add to it, by July 2024, there were other business challenges:

New competitors in market
Influencer marketing becomes expensive
CAC is an all time high

USER INSIGHTS

From my user calling, my hypothesis was that the older program, although accounting for frauds and fake referrals, was felling short because it didn't meet user expectations.

Commission sharing model not enticing for an average user on Probo.
No tracking provided for people invited, loss of trust
"When will I get the reward?", ambiguity

CATEGORYGrowth

MY ROLEStakeholder Buy InDesign ExperimentationUsability TestingHandoff

TEAMDesign - 1Product Manager - 2Engineer - 3

DURATIONOctober 2024 - November 2024

Some recurring anecdotes from user calling:

SOLUTION

After discussing with the product manager and the leadership team, collecting previous knowledge from Referrals program, we broke down the referral journey into manageable milestone metrics:

EXPERIMENT 1

We wanted to find out if promise of money would work better or the social proofing (since there was a trust issue in the older users).

WINNER

Social proofing converted more users than showing monetary rewards.

EXPERIMENT 2

Next, we tested out some more versions to land onto the best converting copy, CTA and interaction.

WINNER

Version 4 emerged as a winner (this was an underdog one, I championed for this version to be tested because the user insights had shown that people wanted something simple, instant and promising)

EXPERIMENT 3

Finally, we tested staggering the referral reward so as to curb fraudulent sign ups and optimise business costs.

WINNER

Scheme 2 emerged as a clear winner. The (0, 10, 40) scheme had successfully slashed the New User CPD by 45% and this scheme was finally made live for 100% of the user base.

CONCLUDING EXPERIMENTS

After all the experiments, we concluded with the best converting copy, interaction and an experience that users seemed to trust.

IMPACT

The revamped referral program not just helped Probo create a strong user acquistion channel, but also gave us lot of love from our user community.

40%

REFERRED USERS / DAU

45%

CPD SAVINGS

1.8X

REFERRALS PER USER

The referral program saw a lot of users posting 'How to' videos on YouTube and becoming micro-influencers!

Learnings

I realised that more rewards don’t always mean more value—especially when they attract bulk referrers gaming the system. True growth comes from nudging real users to take their first step into referrals.
Instant gratification works. Users respond best when the feedback loop between action and reward is tight.
Quality matters more than volume. Conversion and deposit rates are better signals of success than vanity metrics like raw referrals.
Older users are a tougher to win back. Their habits are deeply formed, and moving those metrics needs a strong GTM push layered with personalisation in the customer journey.

Google Sites

Report abuse