I spotted the problem while working on other projects and pulled data from Medallia and Adobe Analytics to build the case for change. When I was asked to bring a project to the team, I knew this was top of my list. I proposed removing 14 of 35 links, phased the work to reduce risk, shipped content and internal linking changes to three live pages, and set up A/B testing infrastructure before organisational restructure paused the broader rollout.
The Problem
The nav had expanded organically over time and ended up with 35 links, inconsistent grouping, and no clear hierarchy. Data from Medallia confirmed that users were clustering almost entirely around a handful of high-value categories: Pay Monthly phones, SIM deals, and Home Broadband. A significant number of links were getting very low engagement and adding noise rather than value.
The harder problem was stakeholder buy-in. Every link in the nav belonged to someone, and nobody wanted to lose visibility. Any proposal that felt like a blunt cull would stall immediately.
My Role
I initiated, led and delivered this project alongside my regular work. I pulled interaction data from Medallia and worked with the team’s data analyst who sourced supporting data from Adobe Analytics. I developed the proposal, led stakeholder conversations, wrote copy briefs, wireframed page changes, and set up post-launch tracking.
Approach
Let the data lead the conversation
Rather than going to stakeholders with a UX opinion, I went with evidence. Showing which links users actually relied on and which were barely touched moved the conversation away from defending links and toward asking which ones were actually earning their place.
Phase the changes to reduce risk
Rather than pushing for a full overhaul in one go, I proposed strengthening internal linking across key landing pages first, so users could still find content even if it moved out of the nav. That would then serve as the evidence base for an A/B test before committing to permanent changes.
Support navigation through content, not just the menu
I addressed the SEO and discoverability concern directly by identifying internal linking opportunities across landing pages and writing briefs for the copywriter to update content on the SIM Only Deals and Tablets and Laptops pages.
Negotiate, don’t force
Where stakeholders pushed back on removing specific links, I didn’t dig in. I agreed to retain contested items with a clear plan to revisit based on post-launch data. Getting most of it live was better than holding out for everything and getting nowhere.
What Went Live
Latest Offers page
I wireframed a new “What offers are you looking for?” section positioned just below the top trending offers, surfacing links to Student Deals, Phone Bundles, Refurbished Phones and Multiline Discounts. The wireframe went straight to design and build without significant changes.
SIM Only Deals page
I wrote a copy brief to clarify the distinction between Pay As You Go SIMs, Pay Monthly Data SIMs and Pay As You Go Data SIMs, particularly around free SIM availability which was creating unnecessary confusion.
Tablets and Laptops pages
I briefed the copywriter to add contextual links strengthening internal linking for both SEO and user navigation, and worked with stakeholders to correct the nav link to point to the main shop page rather than the SEO landing page, bringing it in line with how other device categories were handled.
What was set up for next
With the internal linking changes live, the plan was to run an A/B test showing 50% of visitors the simplified navigation and 50% the existing version. Tracking was defined and everything was in place to run it.
Reflection
The most useful thing I did on this project wasn’t the wireframing or the data analysis. It was reframing the stakeholder conversation. Navigation decisions in large organisations tend to be political rather than user-led, with every team protective of their slice of the menu. Getting people to look at actual usage data changed the dynamic. It’s hard to argue for keeping a link that almost nobody clicks.