Designing for growth inside a locked room
How I reshaped a premium Arabic investment education community inside Circle.so constraints, driving 416% revenue growth in three months.
3 months
3% → 10%
in 90 days
first month
A product with potential and a platform that resisted it
KanzMS launched in October 2025 as a premium Arabic investment education community. A Circle.so community and a landing page existed; the experience told a different story. Navigation was bloated. The brand was inconsistent. Onboarding dropped people into a wall of content with no direction. Conversion was sitting at 3%.
I joined in December 2025 as part of a three-person design team. By February, I was the only designer. The task was not to rebuild, it was to reshape, inside real constraints: no custom layouts, limited styling control, a sidebar that resisted restructuring.
Circle.so as a constraint, not an excuse
Most of the limitations were platform problems, not design problems. Circle.so is built for communities, not brand-driven products: every space renders as a flat list, the builder produces generic pages, and emails carry Circle’s identity by default.
Iframe and custom code injection to build fully branded pages inside the community, bypassing Circle’s native builder. Same approach for transactional emails.
Outsourced forms embedded directly into onboarding posts: a continuous research stream that never interrupts the member experience.
Three decisions that moved the needle
The original page listed what the product offered. The redesign asked a different question: what does it feel like to be a KanzMS member? Neutral dark design system built from scratch, editorial Arabic typography, RTL-first layout. The page stopped selling access and started selling belonging.
The sidebar had become a directory of everything. I rebuilt the information architecture around member intent — what they came to do — not around what the platform could display. Fewer top-level spaces, clearer hierarchy, a restructured onboarding flow that followed naturally. This reduced friction at the exact point where most members previously disengaged.
A weekly quiz system with localStorage attempt-blocking, an SVG countdown timer, and a downloadable score card. It gave members a reason to return on a fixed schedule and created a social proof layer through shared results. 300 responses in its first month confirmed it became one of the most-used parts of the product.
Decisions grounded in evidence
I led research for every feature I built: 1:1 benchmarking against comparable Arabic investment education products, to find where KanzMS over-delivered on low-impact features and missed the high-signal ones.
One finding changed the roadmap: the most-retained users in comparable communities had direct, personal access to an expert. That became 1:1 sessions, a positioning move as much as a design one.
The embedded forms ran throughout, so data shaped each decision rather than explaining it afterwards.
What the numbers confirmed
Measured from December 2025 to February 2026. Marketing contributed to user growth, but the product changes drove the conversion improvement. Growth is ongoing.
“The best design I did here was never visible. It was in the decisions about what not to build.”
Working alone inside a constrained platform forces clarity. Every feature had to earn its place; every workaround had to be invisible.
Conversion is almost never a landing page problem. It is an information architecture problem. When members understand where they are, what they have access to, and what to do next, the numbers move.