Connecting trading education, practice and feedback

Learning without practising

Learners studied theory in one place and practised elsewhere, with little connection between the lesson and the trade.

Practising without understanding

Without feedback, learners could repeat mistakes in a simulator without understanding what to change.

Analysing without a system

Courses, charts, journals and assistants held different pieces of the learner’s progress. Switching between them made it harder to see patterns.

A business that depends on the upgrade

The first level is free. The product earns only when a learner chooses to pay and then stays, so every exit in the payment journey matters.

Competitor research

I reviewed journaling tools, course platforms and trading-psychology trackers. The gap I identified was a connected experience for learning, practice and review; that shaped our positioning and roadmap.

Prototype, then test with users

I prototype new ideas in v0 and Lovable and run usability tests on them, using what people struggle with to remove friction before engineering time is committed.

Funnel data alongside feedback

I combine SQL funnel analysis with user research and feedback to decide what to build. Funnel and billing data showed where people dropped out of the payment journey.

Connect learning, practice and review

Courses, paper trading, the journal, pattern detection and the AI tutor were designed as one system. Every paper trade syncs to the journal automatically, and the AI reads the journal.

Feedback is what turns practice into learning. A learner should be able to practise what they just studied, then see why a trade worked or failed.

Keep real-money trading out of scope

I kept a live trading terminal and links to existing brokers out of scope. Practice runs on a paper account with live prices.

The product's job is to make someone a better trader before they risk money. A terminal would have pulled the team towards execution and away from teaching.

Trade-off: Learners who wanted to place real trades in the same app had to do that elsewhere.

Let learners try the product and progress by assessment

Level 1 is free with no card required. Each later level unlocks only after the learner passes the assessment for the one before.

A free first level lets learners assess the product before paying. Assessments make progression depend on demonstrated understanding.

Treat stale data as a product problem

For the TradingView integration I worked with engineers on data flows, refresh behaviour, latency, and what the app shows when a feed is delayed, stale or fails.

Refresh behaviour and stale-data handling decide whether users trust what they are looking at. A wrong price is worse than no price.

Make billing self-service, failures included

I mapped onboarding, authentication and payment end to end before the build, then owned the Paystack subscription layer: upgrade flow, billing states, failed payments, retries and renewals.

Users needed clear paths to sign up, verify, pay and recover access. I specified failures and recovery alongside the successful journey.

Add a quarterly plan

I created a quarterly payment option alongside monthly billing.

The option was intended to support a longer learning commitment alongside monthly billing.

Build alerts as the reason to come back

I specified email and phone authentication, plus real-time notifications for trading signals and live classes.

These events are time-sensitive. Timely alerts give learners a reason to return; notifications now account for about 70% of engagement.

Structured courses and certificates

Four levels from foundation to expert, each ending in an assessment and a verifiable certificate.

Paper trading

A simulator with a virtual balance, live prices, long and short positions, leverage, and market, limit and stop orders.

Trade journal and pattern recognition

Trades sync from the simulator to a journal, and the product detects patterns in a learner's own history.

An AI tutor

Chat, chart analysis, a scenario simulator and a tutor inside each lesson.

Live market data

TradingView API integration for live forex and crypto price feeds and charting.

Payments and subscriptions

Paystack integration covering card payments, the upgrade flow, billing states, failed payments, retries and renewals, with monthly, quarterly and annual billing.

Sign-up, sign-in and alerts

Email and phone authentication, onboarding with failure and recovery paths, and push notifications for trading signals and live-class alerts.

Help that ships with the product

User FAQs and documentation released alongside each version.

Find what is time-sensitive first

Notifications account for about 70% of engagement. That taught me to identify where timing affects a product’s value and test how reliably it reaches users.

Billing is never finished

I specified the failure paths before the build, and live use still found cases the specs had missed: plan changes that left subscriptions overlapping, and a bug affecting Apple purchases. Both were fixed in a later release.

Fix what moves conversion

Funnel and billing data pointed to the fixes that moved people to paid, and conversion doubled. Those were not always the fixes that were easiest to argue for.

Saying no protects the product

Leaving the trading terminal out was the hardest call and the most useful one. It kept a team of eight building one thing well.