An independent strategy memo on where Binance should invest its next major UX bets — and what to deprioritize.
Binance's product works for the average user and fails for the user at the edge. At 300M users, the edges are where the business lives — and where the next billion will arrive. The mandate is to design for the exception, not the average.
28+ public sources synthesized with Claude as research collaborator — Trustpilot (5,943 reviews), App Store, Google Play, G2, Capterra, Reddit, and Binance's own incident reports. Sources span 2024–2026 to reflect the product as it stands today.
AI accelerated source triage and pattern detection across thousands of reviews. Human judgment owned framing, prioritization, and every strategic recommendation. Where AI synthesis disagreed with my read of the data — most often by treating positive corporate disclosures as resolution rather than as one data point among many — I overrode it. The output is a working test of where AI augments senior design work, and where it doesn't.
A parallel body of work — Designing Intelligence: UX in the Age of AI — explores the principles, archetypes, and prototypes behind how I think about designing with AI: tinkin.xyz/ai-ux-design.
300M+ registered users across 180+ countries, daily trading volume up 18% year-over-year. Yet the majority of new signups have never touched crypto before. Serving them the same interface built for institutional traders creates anxiety, confusion, and churn. The Trustpilot score (1.4/5, 82% one-star) versus the App Store score (4.8★) tells the precise story: the product works for the user it was designed for, and fails for the user it actually has.
Anxious. Curious.
Needs trust, clarity, and guided action.
Power user. Impatient.
Needs raw data density and execution speed.
Deep research across Trustpilot, App Store, Google Play, Reddit, G2, Capterra, and Binance's own incident reports — 2024–2026 only. Each pain point below is paired with its strategic implication: what fixing it actually requires of the company beyond the design team.
Infrastructure collapses exactly when traders need it most.
$283M in compensation is the price of an infrastructure failure that the UI made worse: the app didn't tell users it was failing. Fixing the underlying reliability is engineering's lane — not design's to lead. What design owns is the legibility layer: making latency, queue depth, and order state visible to users in real time, every day, not just during crashes. Trust is built before the storm, not after — and that's a design responsibility even when the failure mode isn't.
Same address format. Permanently different consequences.
When a platform publishes a recovery article for a flow it owns, it has chosen documentation over design. The strategic move is to treat irrecoverable user errors as a P0 metric — counted, reported, owned at the executive level — until the underlying flow is auto-detected, defaulted, and gated. Network selection is the highest-leverage edit in the entire app.
Better — but not good enough.
The "UI Refined" update is a meaningful step. But a binary lite/pro switch still treats complexity as something users opt into wholesale. The strategic move is to tie feature visibility to user behavior, not user choice — Day 1 sees buy/sell, first trade unlocks limit orders, KYC level 2 unlocks margin. Users earn complexity; they don't have to ask for it.
AI replaced humans — and created a worse problem.
Design cannot fix what is fundamentally a refusal to fund human escalation. The 79% auto-resolution stat is real; so is the 21% — and the 21% are precisely the users with money on the line and zero tolerance for friction. The strategic move is to make the tradeoff legible to leadership: every dollar saved on support staff is a dollar lost in trust-driven churn. Design's role here is to instrument the cost, not paper over it.
Fast for 98%. Nightmarish for the rest.
A 98% pass rate sounds like success. At 300M users, the 2% failure mode is six million people locked in verification limbo — and they're disproportionately the users who most need access. The strategic move is to redefine the headline KYC metric from "% approved" to "median time-to-resolution for the worst-decile case." Average-case optimization is what got Binance here. Worst-case design is what gets them out.
Broken, unreliable — and for margin calls, officially not guaranteed.
When the documentation says delivery isn't guaranteed for the most financially consequential notification a user can receive, the company has chosen liability protection over user safety. The strategic move is for design to push for tiered delivery guarantees: critical alerts (margin, security) get push + SMS + email with confirmation — and the SLA is published to users. Anything less is unserious for a platform with leverage products.
Real and fake Binance texts arrive in the same thread.
If a Web3 expert nearly falls for a phishing SMS, the channel itself is the problem. The strategic move is to migrate trust-critical communication into the app — verifiable in-app banners, anti-phishing codes on every official message, and a "verify this message" paste-checker. SMS becomes a fallback, not the default. This is a security investment that reads as a UX investment, which is why design has to lead it.
The pattern behind all 7 pain points: Binance's core product works beautifully under normal conditions. Every failure happens at the edges — during volatility spikes, KYC edge cases, wrong network selections, AI support loops. Designing for the exception, not just the average, is the mandate for this role.
28 SOURCES · TRUSTPILOT · APP STORE · G2 · CAPTERRA · REDDIT
A strategy that doesn't say no isn't a strategy. Below: my force-ranked priority list, with explicit reasoning for what to invest in and what to deliberately deprioritize. Reasonable people will disagree with this ranking. That's the point.
Why first: the only pain point on this list that produces unrecoverable, irreversible user loss. Every day this isn't fixed, real users lose real money to a flow Binance owns end-to-end. Highest leverage, clearest scope — a single flow, not a platform refactor. Success metric: wrong-network withdrawal incidents per million transactions, reported to leadership weekly.
Why second: the single largest driver of Trustpilot's 1.4 score. Fixing the failure path for the worst 2% does more for brand trust than any UI refinement to the happy path. Also the pain point most directly in design's domain — instructions, status, expectations — without requiring infra rebuilds. Success metric: median time-to-resolution for the worst-decile KYC case, not the median or average case.
Why third: compounds with #1 and #2. A user who has just lost funds or been stuck in KYC for 30 days needs a visible, trusted path to a human. Without it, every other fix leaks credibility. The hard part is organizational, not technical — design has to make the cost of dead-ending users legible to the people deciding the support budget. Success metric: time from user-flagged dissatisfaction to human contact, measured at the 95th percentile.
Visible. Expensive. Mostly an infrastructure problem with a UX wrapper. Real fix lives in matching engine and order routing — not the design team's lane to lead. Design's role is instrumentation and graceful degradation, which is contributory, not driving.
Real problem. Backend reliability problem first, design problem second. Design can specify the SLA and surface delivery state, but cannot deliver the underlying guarantee. Sequenced behind #1–3.
High consequence, low frequency relative to #1–3. Mostly a security and carrier-relations problem — design contributes the in-app verification surface but doesn't lead. Worth a quarter of focus, not a year.
Already partially shipped via the June 2025 "UI Refined" update. Diminishing returns on iteration here while #1–3 remain unaddressed. Revisit once the trust deficit is closed; a beautiful onboarding into a system users don't trust solves nothing.
If I could keep only one of these, it would be #1. Network selection failures are the only pain point on this list where a perfectly designed fix produces a permanently better user — every other pain point is recoverable, even slowly.
The seven pain points are the present-tense argument for change. The three priorities above are what to fix in 2026. The five pillars below are what Binance becomes over the following three years — assuming those fixes succeed and the design org earns the right to lead bigger work. Each pillar is a workstream a senior design lead could own end-to-end, and each connects to at least one of the documented pain points.
Buy/Sell. Portfolio view. Simple wallet. That is the entire surface. Trust signals, plain language, zero leverage products visible.
Limit orders unlock. Basic charts unlock. Watchlist appears. Education layer surfaces contextually — never as a tutorial wall.
Full order book. Indicators. Multi-asset views. Customization hands off from Pillar 01 to Pillar 02 at this point.
Margin and futures unlock — gated by both behavior and verification. API access. Bloomberg-grade density, on demand.
The shift from "UI Refined" to earned complexity: Binance's June 2025 Lite/Pro toggle is a meaningful step, but it still treats complexity as a wholesale opt-in. Earned complexity ties feature visibility to user behavior — Day 1 sees buy/sell, first trade unlocks limit orders, KYC level 2 unlocks margin. Users grow into Binance's power; it doesn't drop on them all at once.
One layout for everyone. No customization. You adapt to the product. Leaves power users underserved and beginners overwhelmed.
Infinitely configurable but famously intimidating. Requires weeks of learning. Only viable for professionals with dedicated training time.
Smart defaults for beginners. Gradual unlocking of modules via Pillar 01. Full power-user control once earned. The product adapts at every stage.
Drag-and-drop portfolio view, watchlist, P&L summary, news feed. Show only what matters. Pin top assets front and center.
Choose chart type, indicator set, order panel position. Beginners see simplified buy/sell. Pros unlock full order book and multi-chart views.
Price alerts, portfolio milestones, news triggers — all configurable. Beginners get curated nudges. Pros build complex conditional alert stacks.
Shortcut bar for most-used features — DCA scheduler, convert, earn, send. Users define their own power menu. Zero navigation depth for frequent actions.
AI observes usage patterns for 7 days, then proposes a personalized layout. One tap to apply. Always editable. Always overrideable.
Layout, watchlists, alerts sync across mobile, web, desktop. The personalization investment moves with the user — not stranded on one device.
iPhone ships with a default homescreen — clean, approachable, works out of the box. Every user rearranges it to match how they live. Binance should work the same way: a great default, infinitely personal over time.
Users who customize an interface invest psychologically in it. A personalized Binance dashboard creates switching cost through emotional ownership. The more you build it, the less you want to leave.
Core exchange
Earn on holdings
Leveraged trading
MiCA restricted
Crypto payments
Digital assets
Education + quiz
Personalized nudges
What you see when you tap "Trade" varies dramatically by country. This is the core UX design challenge — and opportunity.
| PRODUCT / FEATURE | 🇺🇸 US Binance.US |
🇩🇪 EU MiCA Zone |
🇬🇧 UK | 🇸🇬 Singapore | 🇧🇷 Brazil | 🇹🇷 Turkey |
|---|---|---|---|---|---|---|
| Spot Trading | ✅ 160+ coins | ✅ 600+ coins | ✅ 600+ coins | ✅ 600+ coins | ✅ 600+ coins | ✅ 600+ coins |
| Futures / Derivatives Up to 125x leverage | 🚫 Blocked CFTC rules | ⚠️ Credits only BNFCRs, no USDT | ✅ Full access | ✅ Full access | ✅ Full access | ✅ Full access |
| Margin Trading Up to 10–20x leverage | 🚫 Blocked US regulations | ✅ Available | ✅ Available | ✅ Available | ✅ Available | ✅ Available |
| USDT Spot Trading World's #1 stablecoin | ✅ Available | 🚫 Delisted MiCA Mar 2025 | ✅ Available | ✅ Available | ✅ Available | ✅ Available |
| P2P Trading 800+ payment methods | 🚫 Unavailable | ✅ Available | ✅ Available | ✅ Available | ✅ BRL + Pix | ✅ TRY pairs |
| Earn / Staking Flexible + locked yields | ⚠️ Limited Staking only | ⚠️ USDC/EURI No USDT earn | ✅ Full range | ✅ Full range | ✅ Full range | ✅ Full range |
| Launchpad / Launchpool New token access | 🚫 Unavailable | ✅ Available | ✅ Available | ✅ Available | ✅ Available | ✅ Available |
| Copy Trading Follow pro traders | 🚫 Unavailable | ✅ Available | ✅ Available | ✅ Available | ✅ Available | ✅ Available |
| Fiat On-Ramp Buy crypto with local currency | ✅ USD / ACH | ✅ EUR / SEPA | ✅ GBP / FPS | ✅ SGD | ✅ BRL / Pix | ✅ TRY |
| Options Trading European-style, USDT settled | 🚫 Unavailable | ⚠️ Limited | ✅ Available | ✅ Available | ✅ Available | ✅ Available |
The strategic implication: A US user tapping "Trade" sees 3 options. A Singapore user sees 9. Same button, same brand, radically different experience. A modular, region-aware design system isn't aesthetic — it's how Binance scales trust across 180+ countries without fragmenting into chaos. The companies that win the next decade of consumer fintech will be the ones whose design systems can absorb regulatory change as a feature, not a fire drill.
Fractional positions starting at cents. One-tap flows that work on 2G. Designed for users whose first financial account is a phone, not a bank.
Financial assistant fluent in Swahili, Pidgin, Hindi, Bahasa — not just English. Explains crypto in cultural context. Voice-first for low-literacy users. Direct response to AI support's English-first failure mode.
Identity verification for users without a passport, permanent address, or credit history. Alternative document flows, local ID support, progressive verification. Direct response to the worst-decile KYC pain point.
Unreliable connectivity isn't an edge case in emerging markets — it's the default. Core flows like sending, receiving, and checking balance work on 2G or worse. Design for the network you have.
In markets without reliable bank transfers, P2P is the only door in. Designing that flow for a first-time user requires trust signals, fraud guardrails, and plain-language guidance the current UI assumes you don't need.
UX flows for tokenized land, savings, and remittances — giving people in emerging markets access to asset classes that have always been out of reach.
Why this is its own pillar: Localization at this depth is a parallel product surface — its own onboarding logic, its own KYC affordances, its own AI behavior. Treating it as an addendum to other pillars is how every Western consumer fintech has historically failed in emerging markets. Binance has the infrastructure to do better.
Understands your goals, risk tolerance, investment horizon, and behavioral patterns. Builds a tailored crypto portfolio strategy — like having a Fidelity adviser, but for Web3.
Continuously monitors your holdings, surfaces opportunities, flags risks, and explains market moves in plain language. Weekly digest. Real-time alerts. No jargon.
From natural language to executed trade. "Buy $200 of BTC when it dips below 60K" — the agent watches, waits, executes. Automates DCA, rebalancing, stop-loss strategies.
Guides new users through their first trade, explains every term in context, builds confidence step by step. The difference between a user who churns in week 1 and one who stays for years.
Reads global news, on-chain data, social sentiment, macro signals in real time. Surfaces what matters to your specific portfolio. Noise filter and signal amplifier.
Tracks cost basis across every trade, calculates tax liability per jurisdiction, exports reports. The most hated part of crypto — solved automatically, in the background.
No one else has the combination: native blockchain infrastructure, 300M users, global market access, and real trading execution. ChatGPT can advise — it can't execute. Robinhood can execute — it lacks crypto depth. Binance is the only platform that closes the full loop.
AI agents are only as good as their interface. The design work is: trust scaffolding (users handing execution to AI need radical transparency), confirmation flows (fast for pros, reassuring for beginners), and explainability UI — always showing why the agent acted. The mockup above uses chat because that's where the industry is today. Chat is the wrong primitive for execution. The right one hasn't been designed yet — and that is the open problem worth a senior design hire. My current prototypes exploring alternatives →
Decentralization isn't just a technical property — it should be a design principle. Binance's UI should be as composable and local as the blockchain it's built on.
A strategy memo without epistemic limits is propaganda. The arguments above rest on public data, and public data has predictable blind spots. Before committing significant resource to any of this, here is what I would want to validate from inside the company.
The next question is what running this actually looks like.
Dedicate the first quarter to network selection auto-detection and gating, end to end. Ship a measurable reduction in wrong-network withdrawal incidents within 90 days. Use that as the proof that the design org can move fast on consequential work, then earn the right to take on #2 and #3. Do not start three workstreams simultaneously.
Two senior product designers, one content designer, one design technologist, one researcher with regional fluency, and direct embed with the relevant eng and PM leads. Fewer designers than the org would naturally allocate, more seniority per head. The work is judgment-heavy and tradeoff-heavy; junior staffing on this set of problems is a false economy.
Wrong-network incidents per million withdrawals. Worst-decile KYC time-to-resolution. P95 time from user-flagged dissatisfaction to human contact. Three numbers, reported to leadership weekly. Average-case metrics — which Binance already optimizes well — would not be on my dashboard. The job is to fix the edges.
Official 2025 EoY report. Daily trading volume up 18% YoY. Yet Binance's Trustpilot score sits at 1.4/5 — the biggest gap between market dominance and user satisfaction in crypto.
Coinbase retains 62% of users vs Binance's 59%. At 300M users, a 3-point gap is millions of churned users annually. Closing it through UX is a direct revenue line.
$145T traded all-time. $162.8B in user balances verified via Proof of Reserves. The infrastructure is world-class. The UX experience should match it.
The World Bank's estimate of people without basic financial services. Binance's infrastructure — 490 coins, 1,889 trading pairs, $1.2B in Earn rewards — could reach them where no bank can.
Futures unavailable in 44+ countries. EU lost USDT access in 2025. A modular design system isn't aesthetic — it's how Binance adapts faster than any competitor.
~1,500 staff (growing 30% YoY) managing 18+ active licenses. Regulatory complexity is real. Design systems that degrade gracefully per jurisdiction are a business necessity.
Whether you're a trader in Tokyo or a first-time investor in Lagos, Binance should feel like it was built exactly for you. The infrastructure exists. The design challenge is making 300M users feel that way — by fixing the edges first, and letting the average take care of itself.
