For 26 months, from March 2024 to May 2026, I worked remotely as a product designer for Muthabr, an EdTech platform based in Muscat, Oman. It serves students from Grade 5 to Grade 12, alongside teachers and the account holders who manage access, across a mobile app, a web app, a marketing website and email.
A lot of my work sat on the short path between "I want to study" and "I'm studying": creating an account, verifying it, choosing a subscription and finding the right activity. This post covers two parts of that path, onboarding and subscriptions, and the research that shaped how I redesigned them.
One note before the numbers. Everything below is research, not business results. It tells you what students needed and how the designs responded, not what the redesign did to revenue.
Starting with where students already learn
Before talking to anyone, I looked at three products students already use: Quizlet, Khan Academy and Duolingo. I wasn't looking for features to copy. I wanted the patterns students would arrive already knowing.
Quizlet lets you find content by subject and get straight to review material.
Khan Academy has a clear course hierarchy, from syllabus to unit to topic.
Duolingo makes progress and rewards visible without getting in the way of the lesson.
Each one mapped to a Muthabr problem: direct access to revision content, a navigable curriculum, and progress that motivates without distracting. Borrowing familiar patterns means students spend less effort learning the app and more on studying.
What 48 students told me
I surveyed 48 students aged 15 to 18, focusing on older secondary-school students because they are the ones preparing for exams and, often, choosing what to pay for.
67% (32 of 48) said complicated onboarding frustrated them.
63% (30 of 48) said they need a clear subscription total before payment.
83% (40 of 48) said they value study organisation and progress.
Survey numbers tell you what matters. They don't tell you why, so I followed up with six student interviews of 25 to 35 minutes each. I asked questions like "What makes you abandon an educational app sign-up?" and "What information would reassure you before paying?"
One answer stuck with me more than any chart:
"If I get stuck verifying my account, I usually try something else."
I grouped the interview observations with affinity mapping, and four themes came out: account access (confusing sign-in versus sign-up, verification dead ends, unclear recovery), subscription choices (unclear final totals, a need for reassurance before paying), learning discovery (grade, semester, subject, unit, activity) and motivation (progress and feedback).
Where the journey breaks
I turned the research into two personas and a journey map. Sarah, 17, is exam-focused and her line on the persona card is "Help me get to what I need to practise." Ahmed, 16, compares resources and manages his own access: "Show me what I'm selecting before I pay."
The journey map made the problem visible. Students arrive curious, become uncertain while exploring, confused at registration, hesitant at verification, frustrated comparing plans and anxious at the price review. Only once they're studying do they feel relieved.
In other words, the emotional low point sits right before payment, and the steps just before it are the ones that decide whether a student ever reaches the content.
Onboarding: design the route back
The registration flow worked when everything went right. The trouble started when something went wrong, which is exactly what that interview quote was describing.
So I redesigned the failure path first:
Sign-in and account creation are separate entry points, so nobody tries to log in to an account that doesn't exist yet.
Errors explain what happened and what to do next.
Verification has a way back. If a code fails, the student can resend it and return to the same step instead of starting again.
The other onboarding problem was who the account is for. One account can involve an account holder, a student and a teacher. I separated account ownership from learner details, so verification, student or teacher details and profile passcodes became distinct steps rather than one long form.
The lesson I took from this applies well beyond EdTech: test your sign-up flow by breaking it. Mistype the code, let it expire, close the app halfway. The happy path is rarely where people leave.
Subscriptions: make the total impossible to miss
I then ran moderated remote usability sessions with eight students aged 15 to 18, covering four core flows: accessing an account, reaching study material, choosing a subscription and connecting with friends.
Some of it worked. Account registration and price review had successful completions, so I kept the basic account and purchase structure and focused changes on the steps that caused uncertainty. Leaving working parts alone is a design decision too, and it kept the changes small enough to evaluate.
What didn't work was understanding what each plan included. Students struggled with subscription access, and the plan options and final payment total needed clearer presentation.
The redesign:
Semester and full-grade options are separated, so students compare like with like.
Savings are visible, not left for the student to calculate.
A subject-selection step comes before payment, so the plan reflects what the student will actually study.
The final total, including any discount, is shown before payment.
That fourth point is the 63% from the survey. But the sessions showed that the number alone isn't enough. A total only reassures people when they already understand what it covers.
Other changes that came out of testing
Testing also showed that the friends view needed a clearer distinction between connections, requests and suggestions. I separated the three and added dedicated ways to find friends through contacts, username or profile sharing.
It also surfaced an opportunity for a referral journey. I designed one that appears after subscribing, keeps the referral code and reward terms together, and tracks balance transfers, with an entry point in settings and a shortcut in the learning view.
What I'd do next
Re-test the recovery paths specifically. I'd run a round focused only on failure states (wrong code, expired code, forgotten passcode) and measure task completion and hesitation.
Widen the sample. The survey and sessions focused on students aged 15 to 18. Muthabr also serves Grades 5 to 8, teachers and account holders, and their needs at the payment step are likely to differ.
Instrument the funnel. Step-by-step analytics on registration, verification and plan selection would show where people drop out at scale, not just in eight sessions.
Compare payment summaries. Once there's enough traffic, an A/B test of the subscription summary would show which presentation helps people decide.
Takeaways for anyone designing sign-up and pricing
Design the failure states before you polish the happy path.
Separate who pays from who uses the product.
Show what a plan includes before you show what it costs.
Keep the parts that already work, so you can tell which changes made the difference.
You can read the full case study, including the personas, journey map and redesigned screens, on my portfolio: Muthabr case study.
I'm a product designer based in Brighton with a computer science engineering background, so I research, design and build. I'm open to product and UX designer roles (full-time, part-time or contract) and to freelance projects, especially EdTech, e-commerce and Shopify. Get in touch.
