This should probably be a blog post, but… some context

The redesign didn’t start with a brief,
it started with a marketing report

In 2021, we redesigned our mobile app, and the web app in 2022. Classic ‘we found product-market fit, now let’s make it better’ energy — and it was the right call at the time. But as Obiex grew, the product outgrew not just our designs, but also our processes, policies, and the way we approached work.

We’d begun extending Obiex into live events when the first real wave hit. Our marketing lead, Ty, was struggling to convert leads — not because his pitch was anything less than impeccable — but because of the onboarding flow waiting on the other side.

We’ll give Ty his flowers — he didn’t come raging at us despite our being the clog (not cog) in his wheel. But he opened up a door we could all see from the corner of our eyes — through every quick fix and patch — yet never dared look into, afraid to find exactly what we did find: a somewhat muddy and dark path.

a lil’ glitch to the flow — just to show the evolution of our onboarding screens… because we can

It was like Pandora’s box, only worse

Problems, problems, everywhere! This is where we should put our problem statement, but…

There were many layers, so we couldn’t possibly write just one problem statement. Like mythical box, the more we pulled, the more there was to pull. So here’s a breakdown of how it all went down.

Layer one: The tug of war!

As regulations grew with our scale, the compliance team needed more user information. But our existing approach of collecting it all upfront added friction to an onboarding flow the marketing team was already struggling with. Compliance became a bottleneck for marketing, and we were caught in the middle — figuring out how to serve two very reasonable, very essential needs.

We became referees in our imagined tug of war

No actual fight broke out — but what a blast it would have been!

Layer two: It’s us, we’re the problem, it’s us

What started as a design problem — scattered Figma files, reviews tracked nowhere, a few detached instances here and there — quickly became a dev problem too. Poor handoff flows, disorganised comments, zero documentation, and code components slowly going stale. Not to mention a serious dark mode problem that was costing us real design time.

we’d built a design system once, and quietly thought we’d arrived.

boy were we wrong!

Layer three: The false positives

We consulted our existing customers and found they had very few issues with the product experience — a proud moment. But problems existed nonetheless, and an even harder-to-see problem was users who churned before they ever became customers. As usual, the feedback we gathered was valuable, and every issue has since been addressed by the teams that own them.

we documented some of their comments, not quoted verbatim, with fictional names.

so how did we solve these problems?

For the tug of war, we turned the rope into a bridge

After rounds of negotiation, we landed on a checkpoint system: collect information only when it’s actually needed. Marketing and customers got the fast, low-friction onboarding they wanted. Compliance got every check it required — gathered the moment it mattered, not a second sooner.

The registration screen went from jarring to effortless

We gave users a chance to explore what Obiex offers before asking for their details — no more checkpoint at the door

Now, let’s walk through some before and after

Back to how we changed our customers’ lives… lol. Starting with what they asked for

Our customers are so great — they don’t ask for much.

and now to things they didn’t even ask for

Aren’t we great?!!

None of this happened on the rust-and-rails of the process we’d been limping along with

The first thing we did was clean house. Variables, tokens, and components rebuilt to scale rather than just to ship — and a Figma structure that treated documentation and change tracking as first-class citizens, instead of the thing you always mean to do later and never do.

We’d share the actual structure, but that one’s staying our trade secret.

jk — we’re writing a proper case study on that. Stay tuned. For now, here’s our design system, just like every other Behance case study shows it.

We’re also building a Storybook for it — it’s a work in progress, but you can view it here

we also worked on the web app, but it’s for a whole ‘nother case study

Three things we walked away with

we got some comments we’ve now taped to our wall of love

the reviews we got were mixed. we also got plenty of pushback — mostly within the team, during stress testing before anything reached customers. But it was all real, and every piece of friction became a useful insight that sharpened what we shipped.

It’s a continuous process — which is exactly why we said ‘almost right.’ We keep growing.

love, and design🌸