Lead Product Designer · 2021 · Web3 / Real Estate
Building trust language for first-time investors — 0 to 70K+ users, later acquired by ARK7
Tens of thousands onboarded early, 70K+ over the following year, and structural patterns that still held when ARK7 acquired Robinland.
Robinland asked first-time investors to put real dollars into tokenized fractional real estate — mechanics most had never heard of, in a market many didn’t trust — with one year to prove enough product-market signal to keep the company alive. As Lead Product Designer I owned discovery-to-invest experience and the research rhythm inside shipping. Constraint: no separate research or polish phase.
01 · The problem
Teaching and asking for money on the same screen
Robinland asked first-time investors to put real dollars into tokenized fractional real estate, with scheduled yields and map-based discovery. Every piece needed explaining at the exact moment someone was asked to fund it. Hiding the complexity would feel safer short-term and dishonest long-term. Explaining everything up front would kill conversion.
The company also had one year to prove enough product-market signal to survive, so research and polish had to happen inside the shipping loop, not before it.
02 · What I found
People stopped where the interface stopped explaining itself
I ran Maze and UserTesting studies alongside releases and paired them with analytics and stakeholder reviews. The expected story was “users don’t understand Web3.” The real pattern was more specific: people stopped at the seam between browsing and investing, on maps without selectable callouts, and at filters that assumed vocabulary they didn’t have.
That changed the job from simplifying Web3 to explaining it in place.
Process
Validate inside the shipping rhythm
On a one-year clock, research couldn’t sit in a separate phase.
1
Maze / UserTesting
Lightweight studies each iteration
2
Ship
Patterns into the live product
3
Analytics
Behavioral signal + stakeholder review
4
Reframe language
Labels and cues where confidence broke
Continuous loop — research lived inside shipping, not before it
03 · Key decisions
Three calls that built trust with new investors
Decision 1
Keep Web3 visible, and explain it in context
Wallets, tokens and yields stay on screen, with contextual language around them so people learn while they invest.
- Instead of
- Masking wallets, tokens and yields behind simpler metaphors, or putting a tutorial wall in front of the product.
- Why
- Masking would have undercut the investors we needed long-term.
- Trade-off
- Every screen had to teach, not just sell.
Decision 2
Organize discovery around “where am I, and what can I do here?”
Map callouts, guided filter labels, and listings with enough context to evaluate without leaving the page. None of it is exotic. It mattered because we applied it rigorously where users had no prior experience.
- Why
- People stalled on maps without selectable callouts and at filters that assumed vocabulary they didn't have.
Decision 3
Keep Invest in the main line of the journey
Invest became a persistent primary action, reachable from discovery, instead of something people had to find at the end of browsing.
- Why
- The seam between browsing and investing was where people dropped off.
04 · Outcome
From zero to 70K+ users, with patterns that lasted through acquisition
Tens of thousands onboarded early and 70K+ over the following year. ARK7 later acquired Robinland.
The acquisition had many inputs, and design was one of them. What I’d defend is that the patterns from the first quarter still carried the product at scale, because they built vocabulary instead of papering over unfamiliarity.
05 · Reflection
Ship foundations that still work after you grow
Looking back, little of the eventual product was new. Contextual labels, inline guidance and Invest in the main line of the journey: those early calls did most of the work through acquisition.
That’s the bar I take into new work. Not the flashiest launch, but decisions that still hold when the user count, and the stakes, get real.