Case study
Helping Bokun.io boost conversions with a website redesign
CRO
UX Design
Messaging
Intro
Bokun, a B2B SaaS booking platform in Iceland, helps small tour operators take payments, manage bookings, and sell experiences through OTAs. After TripAdvisor bought them, Bokun needed to quickly increase website conversions.
As the CRO consultant on this project, I worked with a UX designer, an SEO expert, and a developer. We revamped Bokun’s digital presence, improving usability and messaging, and significantly boosting conversion rates.
Problem
Bokun’s main problem was understanding their customer base. They focused on attracting enterprise clients, forgetting that most revenue came from small tour operators. Because of this, they struggled to align their CRO strategy and messaging with the decision-making process of both small and large B2B clients.
Solution
My approach to solving Bokun’s problem was unique, focusing on Voice of Customer (VoC) research over traditional personas. I found that issues went beyond design and usability. They involved messaging and Information Architecture. So instead of standard UX practices, we used a holistic CRO method that emphasised deep user research (for better positioning). This let us address both design and content problems at once, increasing conversions more than either can alone.
01. Discovery/Research 🎯
To kickstart the project, my team and I sat down with Bokun to plan research. The goal at this stage was to collect enough data to start validating or disproving assumptions.
Stakeholder Interviews:
As the team and I talked to Bokun about their business goals, it became clear a problem with false consensus bias existed.
Namely, assumptions were being made about the problems users face, and this was bleeding into their website (potentially hurting conversions). To get past this, I asked Bokun to connect us with some of their key customers.
User Interviews:
I delegated the task of interviewing users to the UX designer, but wrote the questions myself. This meant we played the roles of facilitator and observer - removing as much bias as possible from the interviews.
By doing qualitative research like this, we discovered that although most users are happy with the product, key features were absent from the website, and several users resorted to hunting for answers in help docs.
User Research:
Although the interview results were interesting, the big risk of getting attached to ideas at this stage is you can end up designing for a handful of outliers (no matter how “representative” they may appear). Now, we need to validate our research.
Because of time constraints and a lack of existing secondary research, I came up with the solution of using a Python script to scrape the internet for reviews, and then turning those reviews into Voice of Customer data.
This is a technique I’ve found it invaluable on CRO and UX projects as it bridges the gap between qualitative and quantitative research. By collecting large amounts of testimonials and analysing them, you can discover patterns you wouldn’t otherwise find from a handful of survey responses or interviews.
Sentiment Text Analysis:
After collecting 1,000+ reviews, I ran them through a sentiment text analysis tool to identify themes. We compared these themes to user interview data, and found a strong parallel. We also factored them into competitor analysis done by my fellow UX consultant.
Analytics Review:
We were starting to gain a clearer image of our users’ mental models - what they feel and what they expect (attitudes). In order to find out what they actually do (behaviours), I ran an analysis in Google Analytics 4 to identify drop-off points and patterns.
As it turns out, most users barely engaged with the Homepage and navigated straight to the Pricing page. Not unusual for a B2B website, but this could suggest poor positioning.
Heuristic Analysis:
I call this step of the process “quick wins” because usability and positioning issues are easy to fix. To find these, I did heuristic analysis on the site. This turned into a 37-page CRO document that breaks down each page and screen by a set of 6 conversion factors.
02. Define/Synthesise 🧠
By this stage, we had a mountain of raw data and some interesting ideas. In order to narrow our focus, we set about grouping all of our data into themes.
User Journeys and Sitemaps:
The UX consultant mocked up a sitemap and I created a User Journey Map. This process emphasised the fact many users were skipping the feature pages, suggesting a lack of interest or possible navigation issues. By improving the visibility and relevance of these pages within the new sitemap, we aimed to increase user engagement and boost conversion rates.
Problem Statements:
With our list of themes, we ran a group session to define our problem statements. This meant stripping away data that didn’t directly relate to the low conversion issue, and focusing on user challenges we could (a) solve, (b) quantify, and (c) explain:
“Users on our platform are finding it difficult to find answers to pre-sales questions, leading to low engagement with content and low conversion rates”
“Users tend to navigate straight to the price page, but then navigate back to homepage, suggesting the value proposition is not clear on the price page”
“Users are not visiting feature pages, meaning critical information is not being factored into purchase decisions, and increasing resistance”
03. Develop / Ideate 🧭
Now we had defined problems, it was time to find solutions. Because of time constraints, I ran a “How Might We” session within a brainstorming session with the team.
How Might We / Brainstorming:
All of our research pointed towards issues in Information Architecture (IA) as the primary obstacle to optimal UX. Users find it difficult to navigate the website, so we decided to focus on restructuring the sitemap and simplifying designs.
Low-fidelity Wireframes:
The UX consultant reworked the sitemap, and I then took this new structure and applied our research to it.
I created low-fidelity wireframes to present information in order of importance as it appeared in user research.
Although many users navigated straight to the pricing page, we knew this was because of confusion and didn’t contribute strongly to conversions. So we redesigned the user journey to emphasise “Start your free trial”, and include more CTAs.
Branding:
Because we already had wireframes in place, I decided mocking up homepage screens would be more valuable. I took elements of the existing brand identity and combined it with new colour palettes to get signoff from the client before moving on.
High-Fidelity Mockups:
As we worked through UI iterations to land on a finalised design with the client, I made several changes:
Aggressively simplifying the pricing page tables
Using lots of product imagery to improve understanding (user research showed many people struggled to grasp how the platform addressed their needs)
Including more trust symbols to increase confidence
Visualising the most frequently mentioned integrations (Viator connectivity is a primary user concern so I made sure to show this clearly)
We also designed for mobile responsiveness, and created a general theme.
Content Strategy:
Branching out of UX and into straight messaging, I took on the role of copywriter for this project - using the offer value equation and layering in goals, pain points and objections from our research.
Many UX designers outsource messaging, but keeping it in-house made sure copy, wireframes, and final designs were tightly aligned. (Offer messaging being the strongest conversion lever).
04. Deliver/Implement ✅
Through continuous collaboration and feedback loops with the client, my team and I refined the design and messaging.
Developer Handoff:
As we shifted towards the final implementation phase, my focus was on ensuring a smooth handoff to the development team (something which - as a web dev myself - I know can make or break a project). Detailed design specifications, annotated wireframes, and a comprehensive style guide were prepared to make this easy.
A/B Testing:
To ensure ongoing optimisation, I designed a series of A/B tests to be implemented post-launch. These included homepage variations, changing the order of booking funnel steps, and changing the placement and wording of key CTAs. These are currently running.
05. Reflection 🏞️
Working on the Bokun project was a great learning experience, especially when it came to understanding the importance of aligning UX with real user needs. We focused heavily on Voice of Customer (VoC) research, which revealed key design and messaging issues that were holding back conversions.
Looking back, I realise we didn’t focus enough on interaction design, and we could have benefited from more prototyping before launching. However, through iteration, we refined the user experience and made the platform much more intuitive.
n the end, we created a product that truly resonated with Bokun's core audience, resulting in a noticeable increase in conversions.