← All product work

ShareChat · Consumer products · Data

Making discovery worth coming back to.

A diverse consumer audience needed more relevant discovery. Ranking improvements depended on both the quality of the experience and the quality of the evidence.

Role
Associate Product Manager → Product Manager
Product scope
Ranking, discovery and experimentation at ShareChat

01 / The situation

Generic ranking wasted attention across a highly diverse 100M+ MAU consumer ecosystem.

Better discovery compounds into retention, inventory and the strength of the monetization system.

02 / Ownership

Where I was involved.

I worked across feed ranking, interest graphs, instrumentation and the experimentation portfolio at ShareChat.

Product manager across feed ranking, discovery, instrumentation and the experimentation portfolio.

03 / The product decision

Improve the feed and the evidence behind it.

Helping people discover something new without losing relevance, while making sure the event data is reliable enough to guide decisions.

Balance exploration with known interests, and make room on the roadmap for instrumentation before accelerating experiments.

04 / The work

How the pieces fit together.

Ranking and discovery were rebuilt around interest graphs. The product question was how to balance familiar interests with exploration across a diverse audience.

During my APM role, exploration and exploitation changes were tested across more than 20 A/B tests. I also negotiated roadmap space for an instrumentation overhaul.

Later, the experimentation portfolio covered ranking, discovery and monetisation. The work combined product changes with a stronger basis for deciding which changes to retain.

A discovery and learning loop
  1. Behavioural signalsImprove event coverage
  2. Interest graphsBalance relevance and exploration
  3. Feed discoveryMake content easier to find
  4. ExperimentsUse results to inform the next change
Reconstructed view of the published product scope. This illustrates the learning loop, not a production architecture or experiment readout.

05 / Results in context

What changed.

The published results include improved feed click-through rate and broader event coverage. Platform size describes the operating context; it is separate from the outcomes of a specific intervention.

Platform context

100M+monthly active users in the platform ecosystem

Scale of the ShareChat environment, not users personally acquired or an experiment cohort.

Reported outcome

6.1% → 8.3%feed click-through rate

Reported alongside 20+ exploration and exploitation A/B tests during the APM role. It is not presented as the result of a single test.

Measurement quality

54% → 91%event coverage

Reported coverage change from the instrumentation overhaul.

06 / The takeaway

What I carry forward.

Experiment velocity is useful only when the measurement is strong enough to tell the team what to change.