Lira Case Study
Shipped

Commanding attention in a market that's designed to ignore us

Impact
$1.29M AUM
Peak assets
52.8M TRY
Project Lira Boost
Company OKX
Role Content Designer
Launch 0→ 1
Market Turkey

1 - Overview

Context

Extreme local market inflation created a severe retention risk as Turkish users regularly withdrew funds from OKX to earn overnight interest at local banks.

Solution

We launched Lira Boost, an automated sweep feature designed to generate overnight interest on idle funds and retain platform liquidity.

My role

I owned the user activation lifecycle and onboarding segments, turning complex financial mechanics and strict compliance constraints into a frictionless, localized experience.

Business impact

The optimized onboarding directly drove product adoption, scaling the feature to $1.29 million in first-month deposits and successfully retaining at-risk capital.

2 - Localizing content standards

A currency localization system to reduce friction in the activation flow.

Rule 01 Turkish Lira

Full name in trust-building contexts.

Where
Onboarding, marketing, first-time education.
Why
Reads as respectful and credible. New users need clarity, not crypto fluency.
"Earn interest on your Turkish Lira."
Rule 02 TRY

ISO code in trading and portfolio contexts.

Where
Portfolio screens, transaction history, balance lists.
Why
Matches global exchange conventions. Traders scan it instantly.
"Balance: 50,000 TRY"
Rule 03

Symbol in alerts and notifications.

Where
Push notifications, toasts, status banners.
Why
Scans in a glance. Saves space in tight UI.
"You earned ₺ 125."
Rule 04 Don't mix

Never combine formats on one screen.

Where
Design QA, the last check before any screen ships.
Why
Mixed formats break trust. Small doubt compounds.
Turkish Lira · 50,000 TRY · ₺ 50,000
How it travels
For international product suites

Same mapping works for USD, EUR, GBP, MXN. Decide the context tier once, then assign the format per surface.

For localization teams

Currency is one example. The mapping logic also applies to dates, units, and handles across markets.

3 - Simplifying product mechanics

Complex financial mechanics often feel opaque to new users. I simplified three backend rules to keep the onboarding experience clear, predictable, and maintain conversion momentum.

01

Bridging the Earnings Delay

Problem
A delay between deposit and visible earnings triggers doubt and early withdrawals.
Decision
Anchor the user with a precise time expectation right after activation.
Content
"Your Lira is all set to earn tonight. Check back after 4.30 pm."
02

Redemptions take 90 minutes

Problem
A 90-minute lock reads as a trap.
Decision
Surface the wait time upfront
Content
"Your funds should be available within 90 minutes."
03

Early redemption forfeits interest

Problem
Users could lose what they earned with early withdrawals.
Decision
Surface the consequnce
Content
"Redeem now to get your funds sooner, or wait 12 hours to keep today’s interest."
01 / 03

4 - Built from local trust signals

Turkish users had lived through repeated currency crises. Before writing a single word, I partnered with the local market team to understand what users actually feared — not what the product assumed they wanted. What I found shaped every value proposition on the screen.

The onboarding screen
Lira Boost onboarding screen showing Wake up to more TRY value propositions
What I found → what it informed
Finding
Both user types — aspirational and cautious — anchored their trust on a high APY that competed with local banks.
This is why Est 41% APY leads the screen — not buried below. It's the credibility signal that earns the first deposit.
Finding
Cautious users had faced economic distrust. Their biggest concern: is the money liquid? Can I get it back?
This is why Redeem easily is a named value prop, not fine print. Liquid access isn't a feature — it's the reassurance that removes the fear.
Finding
Aspirational users wanted deposit and earning without limits — a platform that matched their ambition, not a cap on growth.
This is why No limit earned its own call-out — and why the headline "Wake up to more TRY" promises guaranteed overnight growth, mirroring what local banks had always offered, now automated.
These findings didn't just inform copy — they built a shared understanding across design and product of why each value prop existed. The screen isn't a list of features. It's a direct response to what this market needed to see.

5 - Rewriting content

The originals weren't wrong. They were written for the system, not the person reading them at 9:41 AM.

State 01 · Delighted
×Original
Lira Boost interest accrual completed 9:41 AM
Net earnings credited to your principal balance. View account history for details.
Revised
You earned ₺ 12.50 in your sleep 9:41 AM
Tap to see your new balance.
State 02 · Convenient
×Original
Auto-sweep service notification 9:41 AM
50.00 TRY transferred from funding wallet. Settlement pending standard processing windows.
Revised
We put your funds to work 9:41 AM
Auto-boost moved 50 TRY to earn overnight for you.
State 03 · Anxious
×Original
Lira Boost sweeping initiated 9:41 AM
Funds will return to main balance in 90 minutes pending settlement confirmation.
Revised
Lira Boost delayed 9:41 AM
Need it before 9:00? Redeem now, or contact us. We're here to help.
01 / 03

6 - Localizing the experience

I bypassed a standard English-to-Turkish translation during onboarding, championing a transcreation strategy to rebuild meaning in a highly cautious market.

The premise

Rebuild meaning, not swap words.

Turkish users have lived through three currency crises in a decade. They scan for proof, not polish, and assume marketing first.

Institutional fatigue Platform-loss memory Proof over polish
The recommendation

Pick one story. Carry it everywhere.

Anchor every surface in stronger terms, or in equal terms with less ritual. When users cannot tell which they are in, skepticism defaults to "marketing."

No ambiguity One promise per surface
Two readers

Different scans, same screen.

High-balance users scan for ceilings and liquidity at size. Safety-first users scan for institutional trust and what happens when something breaks.

Calm language Visible recovery
Localization craft

Small choices that compound trust.

Choose 24-hour time for product detail. Use USD when numbers need to feel stable. Lead with plain value, not casual energy.

Deliberate, not decorative

7 - Impact

First-month deposits
$1.29M AUM
Peak assets held
52.8M TRY
Business impact
  • Retained at-risk capital that was being withdrawn nightly to local banks — turning a platform leakage into a growth lever.
  • Onboarding content directly drove $1.29M in first-month deposits, with zero in-product prompts from marketing.
  • Both user types — aspirational and cautious — activated and stayed, confirming the dual-track messaging strategy held under real conditions.
  • The localized copy and trust-first framing established a content system reusable across future Turkish market features.

8 - What I'd do next time

Research depth

Run moderated interviews with both user types separately — cautious earners who used the platform purely for overnight savings, and active traders who kept funds liquid for positions. Each group has a fundamentally different relationship with idle money. I'd want to understand how each reads "Redeem easily" when their stakes are so different.

Content for traders

For users keeping funds on-platform to stay liquid for trading, the value prop is less about earning and more about flexibility. I'd explore whether messaging like "earn while you wait" or surfacing redemption speed more prominently would convert this segment better — and test it separately from the cautious earner voice.

Post-activation signals

I'd track the first 7 days of notification engagement and measure whether cautious users redeemed early after the 90-minute delay message, or whether aspirational users increased deposits over time. That behavioural data would sharpen the next iteration considerably.

Scalability from day one

Document the dual-voice framework, currency localization rules, and trust-first patterns as a reusable content system — not just decisions made for this feature. The next market launch should start with a foundation, not rebuild the same reasoning from scratch.