Summary
Cunard’s site was a mature product with a lot of redundant content and a confusing navigation.
I proposed, up-sold and then delivered a content strategy to restructure and refocus the site
Proposal
I used a why/how/what storytelling structure to bring my stakeholders on board and help them sell the idea to their stakeholders
The proposal was based on a well-proven methodology.
I divided the work into packages. Each ended with a test phase to validate our findings.
As well as outlining the expected benefits (why), methodology (how), I set out the timeline and inputs I’d need (what).
The business readily approved my proposal and so discovery began
Content audit
I used a crawler app to extract a map of the existing Cunard site
Working with Cunard’s Head of Content, we grouped the 200-odd page types found by the crawler into a set of about 85 high level themes.
These were captured in a spreadsheet from where they could be printed onto physical cards for in-person sorts, or uploaded to an online sorting tool for remote sessions in Cunard’s overseas offices
Card sort
Normally a card sort is a quantitative research activity requiring 30-50 participants.
For my research plan, I adapted the protocol to be more of a qualitative process.
I wanted to understand the participants’ thoughts as they grouped the cards. I let them work uninterrupted, making notes and recording the session and then interviewed them at at the end to understand their desisions.
In all 12 sessions were conducted, with a total of 18 participants.
The sample set was a mix of members of cruise customers and Cunard sales, marketing and client-facing staff.
I suspected that there were gaps in the content tree so new pages might be needed, while other pages could be merged or removed completely. I wanted to test these hypotheses by allowing participants to modify the card deck.
This was another deviation from the classic card sorting protocol, but I made the change based on the outcomes I was seeking.
Tree testing
I worked with Cunard’s web team to create a new sitemap combining the best of the ideas from the sort sessions.
A tree test was used to validate our proposals.
Tree testing is a task-based process where participants are shown a model of a site’s navigation and asked to click where they’d expect to find certain content types.
These are slides from my test report.
Two trees were used - one with the new map and one with the existing one as a control.
In most areas, the new structure outperformed the old one, but there were some tasks where the proposal still underperformed.
However the tree test tool provided extensive data on user behaviour which gave us evidence for further improvements
Final structure and next steps
I ran a series of weekly workshops with the Cunard digital team to finalise the structure and create a plan for implementing the changes.
Week 1 - Rework the structure based on the outputs of the tree test and re-test with the new proposal
Week 2 - Review the re-test outcomes and make any final sitemap tweaks
Week 3 - Goup pages into page types with similar content payloads and functional features, and use this to create a set of required templates
Week 4 - Identify net new vs re-used content
Week 5 - Prioritise the templates for implementation and create a migration plan to move the existing content and create new assets and copy
As this was a content exercise, we were limited to re-using existing CMS components rather than creating new ones. However we did capture a wish-list of new components that would help implement our proposals if development resources could be found