## Background: A Redesign Under Acquisition Pressure<\/p>\n
In 2019, FlixBus \u2014 the European intercity bus platform \u2014 acquired Kamil Ko\u00e7, Turkey’s largest intercity bus operator with over 6,000 employees. I joined the team shortly after as UI\/UX Designer, responsible for two parallel challenges: redesigning kamilkoc.com.tr to improve commercial performance, and integrating Kamil Ko\u00e7’s digital processes with FlixBus’s platform.<\/p>\n
The project delivered a **+2% revenue increase** \u2014 significant at Kamil Ko\u00e7’s transaction volume \u2014 and was recognised as **Best User Experience e-Commerce Website** at the ECHO Awards 2019.<\/p>\n
This is a detailed account of how a structured UX redesign approach produced that outcome.<\/p>\n
—<\/p>\n
## Step 1: Diagnosing the Problem Before Designing Solutions<\/p>\n
Before a single wireframe was produced, I ran a comprehensive UX audit combining quantitative analytics and qualitative research. The site was functional \u2014 transactions were happening \u2014 but the data told a different story.<\/p>\n
**Quantitative signals from analytics and heatmaps:**
\n– Significant drop-off at the seat selection step, with users abandoning after the seat map loaded
\n– Measurable cart abandonment on the payment page despite clear purchase intent
\n– A high volume of bookings completing via the call centre, from users who had started on the web and got confused
\n– Mobile conversion substantially below desktop, despite mobile representing the majority of sessions<\/p>\n
**Qualitative signals from user research (8 interviews + usability sessions):**
\n– Seat selection described as “overwhelming” \u2014 too much simultaneous information with weak visual hierarchy
\n– Booking confirmation used travel industry terminology unfamiliar to occasional travellers
\n– Trust signals weak on the payment page: no visible security indicators, no clear refund policy summary
\n– The mobile experience described as “a tiny version of the desktop site” \u2014 not mobile-native<\/p>\n
The diagnosis was clear: this was a conversion problem, not a traffic problem. The users were arriving. The design was losing them.<\/p>\n
—<\/p>\n
## Phase 1: Search & Results (Months 1\u20132)<\/p>\n
**Problem:** The date picker required manual text input on mobile, causing high error rates and form abandonment.
\n**Solution:** Replaced with a native touch date wheel. Added intelligent defaults (nearest weekend pre-selected for leisure routes).<\/p>\n
**Problem:** Search results displayed base prices, but the checkout total was higher \u2014 a trust-destroying inconsistency that had accumulated unintentionally.
\n**Solution:** Enforced full price consistency throughout the funnel. The number shown on results is the number charged at checkout. In follow-up usability sessions, trust comments about pricing appeared immediately.<\/p>\n
**Problem:** Route departure times were displayed as a flat list with no visual differentiation between time bands.
\n**Solution:** Introduced a visual timeline layout with colour-coded time segments (early morning, morning, afternoon, evening, night), making it scannable without reading every row.<\/p>\n
—<\/p>\n
## Phase 2: Seat Selection \u2014 The Biggest Opportunity (Months 3\u20135)<\/p>\n
The seat map was the biggest single problem and the biggest design opportunity in the entire funnel.<\/p>\n
**Problem:** The map displayed all seats simultaneously \u2014 availability, class, and features all visible at once, with a legend that required reading before the map made sense.
\n**Solution:** Progressive disclosure. Default view shows only availability. Tapping a seat reveals its features. A class filter at the top narrows the map before the user even engages with it.<\/p>\n
**Problem:** Window and aisle seat preference \u2014 one of the most common user desires \u2014 required manually scanning every row.
\n**Solution:** Added a single-tap filter: “Window seats only” \/ “Aisle seats only”. This one addition reduced seat selection time-on-task by approximately 35% in usability testing.<\/p>\n
**Problem:** On mobile, the seat map required constant pinch-to-zoom to interact with individual seats.
\n**Solution:** Completely redesigned the mobile seat selection experience: swipeable coach sections, 44\u00d744pt minimum touch targets per Apple HIG, and a “Best available” one-tap option for users who didn’t want to engage with the map at all.<\/p>\n
—<\/p>\n
## Phase 3: Checkout & Confirmation (Months 6\u20138)<\/p>\n
**Checkout changes:**
\n– Reduced required form fields by removing anything not legally required at booking time \u2014 additional passenger details collected post-booking
\n– Added an inline order summary updating in real time as users entered passenger details, keeping booking context visible throughout
\n– Moved trust signals (SSL indicator, booking guarantee, payment partner logos) immediately adjacent to the payment fields \u2014 not to the footer where no one reads them
\n– Replaced generic validation error messages (*”Invalid input”*) with specific, actionable ones (*”Please enter a valid 11-digit Turkish ID number”*)<\/p>\n
**Confirmation changes:**
\n– Redesigned the confirmation screen and email to lead with the three most operationally important details: departure time, departure point, coach and seat number
\n– Added a one-tap “Add to Calendar” action on the confirmation screen
\n– Introduced a clearly separated post-booking ancillary upsell (luggage insurance, meal pre-order) with explicit opt-in framing<\/p>\n
—<\/p>\n
## The Results<\/p>\n
The redesigned experience launched in Q3 2019. Measured against a comparable pre-redesign baseline, controlling for seasonal traffic variation:<\/p>\n
– **+2% revenue uplift** \u2014 significant at Kamil Ko\u00e7’s transaction volume
\n– **Call centre contact rate** related to web booking confusion dropped measurably
\n– **Mobile-to-desktop conversion gap** narrowed substantially \u2014 mobile became a first-class experience, not a shrunken desktop
\n– **ECHO Award 2019** \u2014 Best User Experience e-Commerce Website<\/p>\n
The “Add to Calendar” feature on the confirmation screen \u2014 half a day of design and development work \u2014 reduced a category of post-booking support contact that had been consuming call centre capacity for years. Small, well-placed interactions compound over time.<\/p>\n
—<\/p>\n
## What Made the Difference<\/p>\n
**Research before redesign.** Four weeks of diagnosis \u2014 analytics, heatmaps, interviews, content and flow audit \u2014 before proposing a single solution. The redesign addressed confirmed problems, not assumed ones.<\/p>\n
**Funnel focus.** We could have started with the homepage (most visible) or the blog (most neglected). We started where the commercial loss was happening. High impact, directly measurable, easy to defend.<\/p>\n
**Stakeholder alignment before design.** A workshop with commercial, customer service, and engineering leads before the design phase meant no surprises in reviews. Everyone had shaped the problem definition; no one was encountering the solution cold.<\/p>\n
**Iterative usability testing.** Two rounds of unmoderated testing (Maze) and one round of moderated sessions during the seat selection phase. The window\/aisle filter insight came directly from watching a user scan the seat map in a moderated session. Without that session, we’d have shipped a better-looking version of the same problem.<\/p>\n
—<\/p>\n
## Lessons That Apply Beyond This Project<\/p>\n
**Conversion optimisation is design work.** The biggest gains here came from structural redesign: how information was presented, how tasks were sequenced, how trust was built. That’s UX, not just A\/B testing button colours.<\/p>\n
**The call centre is a UX research instrument.** The volume and topic distribution of support contacts is one of the richest available data sources. Every category of call represents a design failure somewhere in the user journey.<\/p>\n
**Pricing consistency is a trust mechanism.** Any gap between the price shown in search results and the price charged at checkout destroys trust at the exact moment it’s most needed. This sounds obvious \u2014 and yet it’s one of the most common issues I encounter on e-commerce and booking platforms.<\/p>\n
—<\/p>\n
*Related reading: [UX Strategy That Sticks: Aligning Design Decisions With Business Outcomes](\/blog\/2026\/06\/16\/ux-strategy-aligning-design-decisions-business-outcomes\/) \u00b7 [Information Architecture in UX Design](\/blog\/2026\/06\/16\/information-architecture-2026-how-to-structure-digital-products-that-scale\/)*<\/p>\n
*If your e-commerce or booking platform is losing conversions you can’t fully explain, a structured UX audit is usually the fastest path to answers. [Get in touch.](https:\/\/www.gokhanmeric.com\/#contact)*<\/p>\n","protected":false},"excerpt":{"rendered":"
How a structured UX audit and iterative redesign of kamilkoc.com.tr delivered +2% revenue and the ECHO Award 2019. The specific design decisions across search, seat selection, and checkout \u2014 and the lessons that apply to any conversion-focused redesign.<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1,8],"tags":[33,29,25,30],"class_list":["post-28","post","type-post","status-publish","format-standard","hentry","category-ai-engineer","category-case-studies","tag-information-architecture","tag-product-designer-australia","tag-ui-ux-architect","tag-ux-design"],"yoast_head":"\n