Portfolio
Sales Landing Pages — From Behavioural Evidence to Better Conversion
High visitor numbers were not translating into transactions. With a bounce rate around 90% and conversion at 0.8%, I examined analytics, heatmaps and internal commercial reporting to understand where the landing pages were falling short. My work connected the diagnosis to page structure, stakeholder alignment and subsequent experiments.
Process
Phase 01
Read the behaviour alongside the business
I compared traffic analytics, Hotjar heatmaps and internal reporting to form hypotheses about orientation, missed information and the route to conversion. Using these sources together helped turn observable friction into questions the redesign could address.
Phase 02
Build a clearer decision sequence
I reworked the sequence in which visitors encountered the proposition, benefits and calls to action. Wireframes established that hierarchy; colour and visual emphasis supported it. The design hypothesis was that clearer information and more credible guidance would help visitors decide.
Phase 03
Carry the reasoning into implementation
Stakeholder reviews refined the proposal before I prepared the implementation handoff across product variants, screen sizes and light and dark modes. This carried the intended hierarchy into the conditions under which the pages would actually be delivered.
Phase 04
Let rollout produce the next questions
A split-traffic rollout in one market informed wider deployment. A subsequent series of A/B tests explored CTA colour, benefit copy, language-switch placement and direct sales contact. I developed the hypotheses, analysed the tests and translated the results into recommendations for management.
Methods
Combine complementary evidence
Analytics establishes scale, heatmaps suggest behavioural questions, and commercial reporting supplies business context. Each source contributes something different.
Design around the visitor’s question
The page sequence was organised around what visitors needed to understand about the offer, its value and the next step.
Make the proposition testable
Each variant addressed a defined question. Statistical analysis informed the recommendation, with interpretation still tied to the experiment’s context.
Keep the choice intelligible
Clear benefits and understandable actions guided the work. Visitors should be able to make an informed choice.
Reported conversion rose from 0.8% to 2%, an increase of 1.2 percentage points, or 150% relative to baseline. Click-through improvements reached up to 20%, gross revenue increased by 8%, and the bounce rate declined. Follow-up A/B tests addressed individual refinements. These figures describe the project’s reported results, rather than establishing the isolated contribution of any one design change.