How Phenomenon Studio Redesigned a Broken EdTech Platform, and Cut Learner Drop-off by 47% in Twelve Weeks

By Iryna Huk — Project Manager Lead, Phenomenon Studio February 18, 2026 EdTech LMS Redesign Brand Identity

Key Takeaways

• EdTech platforms that fail do not fail because of missing features — they fail because the learning experience itself is cognitively exhausting. LearnBridge had 14 features its competitors lacked and still lost 61% of learners before course completion.
• Working with a services web design and development partner that prioritizes behavioral data over feature parity is the difference between a redesign that works and a redesign that looks better but produces the same outcomes.
• In our internal benchmarking of seven EdTech platform engagements at Phenomenon Studio (2021–2025), platforms rebuilt with a user-flow-first design system reduced learner support tickets by an average of 38% within 90 days of launch.
• LearnBridge’s course completion rate went from 39% to 72% — a 33-percentage-point improvement — in the first 60 days after the redesigned platform went live. Not one new feature was added in that period

LearnBridge after the Phenomenon Studio redesign, learner dashboard, course progress UI, and instructor portal. Delivered in 12 weeks.

There is a category of digital product problem that is particularly hard to diagnose because it looks like a content problem, or a marketing problem, or a product-market fit problem, when it is actually a design problem. LearnBridge was living inside that category when they came to us in mid-2024. A corporate learning platform targeting mid-size companies in Central and Eastern Europe, they had strong content partnerships, a committed sales team, and a churn rate that was quietly destroying their unit economics. Learners who enrolled in courses were abandoning them at an average of 61% before completion. Corporate buyers were not renewing. And the internal team was stuck in a loop of adding features to solve a problem that features had not caused.

We have been in this situation enough times to know that the answer is almost never found in the feature backlog. We knew from the first conversation that what LearnBridge needed was not more product, it was a completely different understanding of why their users were leaving. Everything we built over the following 12 weeks grew from that starting point.

Case Study Snapshot, LearnBridge LMS Redesign

−47% learner drop-off rate at 60 days post-launch

72% course completion rate, up from 39% pre-redesign

12wk discovery to full production deployment

Client: B2B corporate learning platform, 40-person company, operating across Poland, Romania, and Czech Republic. 280 active corporate clients at engagement start.

Problem: 61% learner abandonment before course completion. Corporate client renewal rate declining quarter-over-quarter for 18 months. Internal team had added 14 new features in that period with no measurable impact on retention.

Scope: Full UX audit and behavioral data analysis, complete brand identity redesign, dashboard UI design for learner and instructor portals, ReactJS front-end rebuild, Node.js API optimization, and a Figma-based design system for internal handoff.

Starting With Behavioral Data, Not Stakeholder Opinions

In my experience running complex platform redesigns, the most dangerous input into a design process is confident stakeholder intuition that has not been tested against user behavior. LearnBridge’s leadership had strong theories about why learners were leaving, the interface felt “corporate,” the mobile experience was poor, the notification system was too aggressive. All of these turned out to be partially true. None of them turned out to be the primary driver of abandonment.

We conducted a structured behavioral analysis of 1,200 learner sessions from the three months preceding our engagement. We cross-referenced session recordings with the platform’s completion data and ran structured exit interviews with 24 learners who had abandoned courses in their first week. What we found was a specific pattern that appeared in 68% of abandonment cases: learners who could not see their progress in relation to the full course scope left within the first two sessions. Not because the content was bad. Not because the interface was ugly. Because the cognitive cost of not knowing how far they had to go was greater than the motivation to continue.

This is a well-documented phenomenon in learning psychology, the “progress visibility effect.” What was not documented was its specific manifestation inside LearnBridge’s interface: a course navigation structure that showed learners only the current module, with no persistent view of overall course structure or completion percentage. Fixing this single interaction pattern was responsible for approximately 60% of the eventual drop-off reduction. Everything else we did contributed the remaining 40%.

Question → Direct Answer

What is the single most impactful thing an EdTech platform can do to reduce learner drop-off?

Make progress visible, persistent, and meaningful. Not a percentage bar hidden in a sidebar, a genuine structural view of where the learner is in relation to where they are going, displayed every time they open a course session. Our data across seven platform engagements shows that persistent progress visibility reduces first-week abandonment by an average of 28 to 35% without any other changes. It is the highest-return design intervention available in the EdTech context, and it is underimplemented in approximately 80% of the platforms we have audited.

“The LearnBridge team had spent 18 months building features to solve an engagement problem. When we showed them that 68% of their abandonment cases traced back to a single navigation pattern, one that could be redesigned in a two-week sprint, there was a long silence in the room. Not because the finding was devastating. Because it meant the solution was actually within reach. That realization changes everything about how a team approaches a redesign.”

Iryna Huk — Project Manager Lead, Phenomenon Studio  |  February 2026

The Brand Identity Work: When a Design System Is Also a Trust Signal

LearnBridge’s existing visual identity had accumulated three years of inconsistency. There were four different button styles across six sections of the platform. The typography system used five typeface weights with no consistent hierarchy. The color palette had expanded from an original two-color brand into an eight-color collection of one-off decisions made under deadline pressure. None of it was wrong individually. Together, it communicated an unconscious message to learners: this platform does not take its own presentation seriously.

We rebuilt the visual identity from the component level up, working with a brand identity agency methodology that starts with brand strategy before touching a pixel. What does LearnBridge stand for in the learner’s experience? The answer, earned confidence, became the organizing principle for every visual decision that followed. Colors that signal progress rather than authority. Typography that is warm and readable at low cognitive load. Motion design that rewards completion rather than demanding attention.

The resulting design system, delivered as a Figma component library with documentation, gave LearnBridge’s internal development team a tool they had never had: a single source of truth for every UI decision. In the six months following our handoff, the internal team shipped four new feature modules using the design system with zero visual inconsistencies introduced. That is the long-term value of a properly built brand identity: it makes every future design decision faster and more consistent, whether or not we are in the room.

⚠ Six Mistakes EdTech Teams Make When Rebuilding a Learning Platform

1. Adding features before diagnosing the behavioral cause of churn. Feature additions feel productive. They rarely solve retention problems rooted in UX friction. Diagnose first with behavioral data, then decide whether a new feature or a design change is the right intervention.
2. Treating mobile as a responsive afterthought. In our research, 62% of learner sessions happen on mobile. Building desktop-first and then “making it responsive” produces a mobile experience that technically works and practically discourages continued use.
3. Hiding progress from the learner. Course completion data is the learner’s primary motivational anchor. If a learner cannot see where they are in relation to where they are going, they will fill that uncertainty with pessimism and disengage. Persistent progress visibility is not a feature — it is a foundational UX requirement.
4. Skipping the brand identity rebuild when the platform feels outdated. A new front end built on an inconsistent visual system produces a platform that looks partially improved and entirely untrustworthy. Brand and UI must be rebuilt together or the result reads as a renovation, not a redesign.
5. Over-engineering the notification system. EdTech platforms frequently use notification volume as a substitute for engagement strategy. Learners who receive more than three automated notifications per week show a 2.4× higher unsubscribe rate in our data. Fewer, smarter notifications outperform volume every time.
6. Defining MVP scope by feature count rather than by user journey completion. What is MVP in software development for an LMS? The smallest set of interactions that allows a learner to complete one full course with no external assistance required. Everything else is phase two.

Dashboard UI Design for Two Completely Different User Types

LearnBridge served two primary user types whose needs were almost entirely non-overlapping: corporate learners who wanted to complete assigned training as efficiently as possible, and L&D managers who needed to track team progress across multiple concurrent programs. The original platform had tried to serve both from a single dashboard structure, which meant neither user group was well served.

We designed entirely separate dashboard experiences for each user type, unified by the same design system but organized around completely different decision hierarchies. The learner dashboard centered on the single most important question in a learner’s day: what do I need to do next, and how long will it take me? The L&D manager dashboard centered on the organizational view: which teams are on track, which are falling behind, and where do I need to intervene?

This separation sounds obvious. Most EdTech platforms implement it partially, a few different views within the same structural template rather than genuinely different information architectures designed for genuinely different cognitive tasks. The difference in user satisfaction between a partial role separation and a genuine one is not small. LearnBridge’s instructor portal NPS went from 31 to 67 in the first 60 days post-launch. That 36-point jump was not driven by new features. It was driven by the fact that L&D managers could finally see what they actually needed to see without navigating through information designed for someone else.

The Technical Rebuild: ReactJS, Performance, and the Mobile-First Decision

We rebuilt the LearnBridge front end in ReactJS, with a component library that served both the learner and instructor portals from the same shared codebase. The previous platform had been built on a PHP/jQuery stack that was technically functional but had accumulated seven years of accumulated technical debt, conditional logic embedded in view files, CSS written without a system, and a JavaScript bundle that took 8.4 seconds to load on a mid-range Android device on 4G.

The new platform achieved a Lighthouse performance score of 91 on mobile at launch. The primary technical decisions that produced that number were predictable loading architecture (critical CSS inlined, non-critical JS deferred), lazy-loading for video lesson content with a thumbnail-first strategy, and a service worker implementation that cached completed lesson data for offline access, a feature that turned out to be meaningfully used by learners commuting on public transport in the Polish and Romanian markets.

We also implemented a hybrid app approach for the mobile experience, wrapping the progressive web app in a lightweight native shell to enable push notification delivery on iOS — a technical choice that sounds incremental but had an outsize impact on the notification engagement rate, which improved from 14% (web push) to 41% (native push).

LearnBridge Before and After: Twelve-Week Redesign Results

*Data sourced from LearnBridge’s Mixpanel analytics, Intercom support volume reports, and Phenomenon Studio’s post-launch monitoring dashboards. Baseline calculated from the 90-day period immediately prior to redesign launch.

What “Done” Actually Looked Like at Week Twelve

At the end of week twelve, LearnBridge had a production-ready platform running on new infrastructure, a complete Figma design system their internal team could extend without calling us, a brand identity document that codified every visual decision we had made and explained the strategic rationale behind it, and a post-launch monitoring setup that gave their product team visibility into learner behavior patterns in real time. We also handed off a prioritized backlog of phase-two improvements, ranked by behavioral data from the first weeks of live usage rather than by stakeholder preference.

I want to be specific about that last point because it is where most agency engagements end and where a genuinely useful partnership continues. A phase-two backlog built from live usage data is fundamentally different from a feature wishlist. It tells the product team where real users are experiencing friction right now, ranked by frequency and severity, with design recommendations attached. It turns the end of a design engagement into the beginning of a data-informed product development cycle. That is what we aim to leave behind at Phenomenon Studio, regardless of the industry or the platform.

Phenomenon Studio’s approach to platform redesign, from behavioral audit to branded design system to production deployment.

Frequently Asked Questions

How does Phenomenon Studio approach a full platform redesign without losing existing users?

We use a phased migration strategy: new UI components are tested in parallel with the existing interface during a beta period, and active users move to the new experience through opt-in before the full cutover. For LearnBridge, this meant zero forced migrations, 94% of active users self-selected into the new interface within three weeks of beta availability. No mass confusion, no support spike, no revenue impact from the transition itself.

What makes EdTech UI UX design different from general SaaS platform design?

EdTech users are frequently in a state of active cognitive effort, they are learning something new, under time pressure, and the cost of interface confusion is disproportionately high because confusion breaks the learning state entirely. This means EdTech UI design must prioritize cognitive clarity above visual ambition. Every animation, every modal, every notification has to justify its existence by making the learner’s task easier, not more visually engaging.

Does Phenomenon Studio offer white-label web development services for EdTech operators?

Yes. We build fully branded LMS shells, custom dashboard UI, instructor and learner portals, and complete Figma design systems for platform operators who need a polished product without building an in-house design and engineering team. The white-label approach works particularly well for EdTech companies entering new verticals or geographic markets where brand differentiation matters from day one.

How important is mobile in EdTech platform design, really?

More important than most teams budget for. In our research across seven EdTech platform engagements, 62% of learner sessions happen on mobile devices, yet only 23% of the platforms we have audited were built with a genuinely mobile-first interaction model. Responsive layout is not enough. Touch targets, offline content access, and notification timing all behave differently on mobile and require deliberate design decisions that desktop-first teams consistently skip, with measurable consequences for completion rates.

IntelligentHQ

Founder Dinis Guarda

IntelligentHQ Your New Business Network.

IntelligentHQ is a Business network and an expert source for finance, capital markets and intelligence for thousands of global business professionals, startups, and companies.

We exist at the point of intersection between technology, social media, finance and innovation.

IntelligentHQ leverages innovation and scale of social digital technology, analytics, news, and distribution to create an unparalleled, full digital medium and social business networks spectrum.

IntelligentHQ is working hard, to become a trusted, and indispensable source of business news and analytics, within financial services and its associated supply chains and ecosystems