Cutting Time to Value with a Passkey-Led Onboarding

Shiga is a cryptocurrency payments fintech building three connected products: ENTA, a consumer wallet for sending, receiving, and converting crypto and fiat; FORGE, a corporate banking platform for business accounts; and a Telegram Mini-app that brings the same primitives into a chat-native surface. The company is backed by Tether and is a member of the Visa Accelerator program.

This case study covers the individual onboarding flow. The business onboarding has its own case study, since the constraints, persona, and KYC weight are different.

RoleProduct design, UX writing
DurationQ1 2026 · 1 week
IndustryFintech, Cryptocurrency, B2C
ForShiga Digital
Team1 × product designer
1 × product manager
3 × engineers
Challenge

The old flow asked users to do too much before they saw the product. Sign up with email and password, complete a profile, finish KYC, then land on the dashboard. Users were churning inside KYC, before ever seeing send, receive, or swap. The funnel review with engineering confirmed it: a meaningful share of new accounts never made it past identity verification. Support tickets carried the same message in plainer language, with users describing fatigue and confusion before they had ever moved a dollar.

There were two problems sitting on top of each other. The authentication primitive was old-fashioned for the category. Wallets like Rabby, Rainbow, Phantom, and Fuse have been moving to passkeys, both for the security model (no password to phish, no seed phrase to lose) and for the recovery story. Shiga's email and password flow looked behind the curve to the audience it was trying to win. The second problem was structural. KYC sat between the user and the product, even though most of what KYC unlocks (fiat onramp, offramp, currency conversion) is not what a user wants on day one. Stablecoin send, receive, and swap can all run on a self-custodial profile with no identity check. The flow was treating regulated and unregulated features as one gate, when they are two.

Solution & Impact

I rebuilt the flow around two ideas. Sign-in becomes a passkey. KYC moves out of the front door and onto the dashboard.

A new user creates a profile with their email, registers a passkey, and lands on the dashboard. From there, send, receive, and swap of stablecoins are immediately available. The features that need KYC (fiat onramp, offramp, conversion, higher limits) are present in the dashboard but visually locked, with a small affordance that explains what each one does and what completing identity verification will unlock. The dashboard, not the onboarding flow, carries the compliance nudge. Users meet the product first, then meet the KYC requirements in the context of the specific feature they want to use.

<1 minCrypto-only user reaches the dashboard
<4 minCrypto & fiat user completes KYC and reaches value
ClearerGoal-driven KYC framing, tied to the feature it unlocks
Behind the Scenes

Discovery

Insight into the problem, researching customers and understanding their needs.

Activities

  • Funnel review with engineering, sign-up to first transaction
  • Support ticket audit for onboarding-related complaints
  • Competitive scan of authentication and onboarding patterns in Rabby, Rainbow, Phantom, and Fuse
  • Stakeholder interviews with product, engineering, and compliance

Key findings

The drop-off was not a UX detail problem; it was a sequencing problem. Users were leaving because the product was withholding value behind a compliance step that, for a meaningful slice of users, was not even necessary on day one. Crypto-native users wanted to move stablecoins. Crypto and fiat users wanted ramps but were willing to invest in KYC if they understood what they were getting. Treating both groups as the same flow made both worse. Competitive teardown showed the wallet category had already moved on from passwords. Passkeys had become the default among the wallets serious users compare Shiga against.

Define

Narrowing the focus, crystallizing the problem.

Activities

  • Persona-aware flow split: crypto-only vs crypto and fiat
  • Authentication primitive change: passkey replaces password
  • Feature accessibility map: what is available pre-KYC, what requires it, and what the upgrade path looks like

Focus area

  • A passwordless sign-up that lands the user on the dashboard within a minute
  • A dashboard that shows the full product surface from day one, with locked states for features that require KYC
  • KYC framed by the value it unlocks, not as an abstract compliance step
  • A clean upgrade path so the same dashboard works before, during, and after verification

Develop

Working on potential solutions.

I ideated end-to-end flows in Claude Design to move fast and pull stakeholder feedback in early, then promoted the strongest path into Figma for high fidelity. Working in a one-week sprint with a product manager and three engineers meant the design needed to be defensible by the second day, so most of the iteration happened on the flow shape rather than on individual screens.

The design decisions that mattered most were the locked-state pattern on the dashboard and the passkey enrolment moment. The locked state had to advertise the feature without baiting and switching the user, so each locked tile shows the action verb (Buy crypto, Convert, Withdraw to bank) with a short note on what completing KYC unlocks. The passkey enrolment had to handle device transfer and recovery cleanly enough that a user signing up on mobile and returning on web did not feel locked out of their own account.

Interactive Prototype

Clickable prototype from the Claude Design ideation stage. Tap through to feel the flow.

Deliver

Finalizing the solutions that work and developer handoff.

The redesign shipped at the end of the one-week sprint. Engineering picked up the work via a single Figma file with linked flows for both personas, and the QA cycle ran in parallel with copy review.

Learnings

Takeaways that underscore a growth process for me.

  • Time to value beats completion rate at the top of the funnel. A user who sees the product in 60 seconds and verifies later is worth more than a user who completes KYC and never logs in again. The right metric for early onboarding is not "did they finish what we asked of them," it is "did they reach value."
  • Passkey is a category signal as much as a security feature. Picking the same authentication primitive as Rabby, Rainbow, Phantom, and Fuse mattered for trust before the user ever touched the product. The category had moved; staying on passwords was its own message.
  • Compliance can live in the product, not in front of it. The dashboard carrying the KYC nudge, with each locked feature explaining the trade, did more to drive verification than a forced flow ever did. Users who chose to verify did so because they wanted the feature on the other side, not because we asked them to.