Customer Support Product Strategy
Delivering a Self-Service Customer Support centre in 8 weeks using structured content.
Customers had trouble building their bikes
Customer satisfaction data, HotJar and a constant stream of unnecessary warranty cases told us customers were struggling. We produced great support information but lacked the resource to manually publish and maintain this information at scale.
My prototypes only highlighted the complexity
We wanted to link the support information directly to the customer’s bike but every prototype I made was too complex and sprawling to build. Conservative estimates for design, development and publishing came out at 6 months. It was Tom, our gifted product owner had the idea of redeveloping the folder structure to unlock content automation allowing us to build once, publish everywhere.
Structuring content means mapping everything first
I’ve spent my first 2 years at Canyon designing our bike launch content strategy. So not only did I know the where, what and why of our support content, chances are I probably briefed and edited the later stuff. This made categorising, mapping and deleting content types time consuming but easy.
Learning about labels
Acquired knowledge is great but not many of the design team have worked so closely with the product. Some newbies lost hours watching video content, reading manuals and re-writing titles because basic meta descriptions was missing from much of the early content.
Super charged micro copy
I worked closely with our UI designer to streamline the component design and remove unnecessary content blocks helping us to reduce our translation costs (we launched globally in 14 different languages).
Automatic for the people
We ended up with a streamlined database of content types to decide how the content should be presented (in a video player, article or download) and product codes to decide where that content should show up. This was handed over to our merchandisers to plug into the new UI and spin up 32 unique product support pages overnight. Score.
Accidental content win
A cute little by-product of this project was the total annihilation of useless support FAQs I inherited from the old site. Devoid of meta descriptions and mostly useless to the customer I was able to delete the worst and funnel the important ones into easy to find articles.
8 weeks end to end
What was previously scoped at 6 months ended up taking two. After 12 months of reporting, we went back again, to improve on the original design and guide more customers to best-in-class support content.