Rebtel
Prepaid Top-Up: designing flexibility without cognitive overload
Migrants sending prepaid recharge to family needed different things depending on who and why. I designed a system that offered that flexibility without forcing users to compare everything at once.
- Role
- Senior Product Designer
- When
- Feb 2019 – Aug 2021
- Also on Behance
- Original deck ↗
Want to ask about this project?
AI answers grounded in Anshula's CV and portfolio.
Rebtel wanted to become a one-stop shop for the things migrants actually need: international calling, top-up, remittance, and had identified prepaid top-up as the next logical extension. But "add a top-up feature" isn't a design problem on its own; the real question was what that feature needed to flex around.
The problem
Sending a recharge isn't a single, predictable transaction. It changes shape depending on who it's for and why. Existing top-up products in the market (PayTM, PhonePe, Tez) were built around single-purpose flows, useful as reference points, but none of them were solving for Rebtel's specific case: a sender composing something closer to a care package for someone they may not see often.
The constraint
This was Rebtel's first attempt at this kind of feature, so there was no internal precedent to build from. No single plan composition fit most people. Senders wanted very different combinations of data, minutes, and SMS depending on who they were sending to.
The tension, and the call I made
The first version of the design reflected survey findings by offering a range of bundle options up front. But usability testing surfaced a real problem: users experienced high cognitive load trying to scan and compare all those different offers and bundles at once.
That finding led to the actual solution: an expandable card system paired with contextual recommendations. Instead of presenting every option simultaneously, the interface surfaced what was likely relevant based on the recipient and prior behavior, with more detail available on demand rather than all at once.
Approach
I led competitive and market research, then ran a user survey and interviews to understand behavioral patterns around prepaid top-up: spend areas, frequency, and what senders were actually willing to do to get money and connectivity to someone they cared about. That research produced three user personas capturing different sender contexts, which grounded the initial design in real, differing needs.
What shipped, tied to outcome
- An expandable card-based top-up flow with contextual recommendations → drove 15–30% adoption within 6–12 months of launch.
- Contextual recommendations surfaced based on prior sending patterns, so returning senders didn't have to reconstruct the same choice from scratch each time.
- Inline plan details shown one option at a time rather than all at once, directly addressing the cognitive load finding from usability testing.
What I'd revisit
The first version underestimated how much the freedom to choose from many bundle combinations would cost users in decision effort. That only became visible once real users tried to use it. It's a useful reminder that solving for "people want different things" and solving for "people can easily get the thing they want" are related but not the same problem.
At a glance
- Improved Feature Adoption by
- ↑0%
- Reduced Cognitive Load by
- ↓0%
- Markets Served
- ↑0+
Want to talk through this kind of work? Let's talk.
More work
