Building a multi-platform entertainment search for Orange entertainment services
Friday, June 10
15:15 – 16:00
UX Designer, Orange France
The initial brief was to redesign, across all platforms (TV, web, applications), the search experience on Orange entertainment services: TV channels, TV on demand, Video on demand, Games, Music, Offers. Seizing the opportunity presented by the implementation of an « elastic search » search technology on the television platform, the project was in fact tackling the breaking down of very siloed departments within the organisation, as each platform and previous technology had been handled separately both form the marketing and design point of vue.
Our main focus was to aligning some 5 different marketing teams, 3 design teams and 4 development teams around the core concept of what search experience means, feels and should behave from the user’s point of view. In addition, taking into account the different teams’ strategies, business models and legal constraints, colliding roadmaps led us to introduce a participative, iterative ad-hoc method with regular public result sharing.
Search activity is an extremely focused activity that is best experienced when the lowest cognitive effort is met by the lowest interaction cost. The design concept is based on 4 fundamental aspects:
- a drastic reduction in time « spent and perceived » for a query; this is attained by devising a new autocomplete architecture from the first letter entered and striving to serve the answer on a one-time basis for the majority of cases. Unmet queries led us to introduce a new category of answer : 1 result /0 content
clarified result reading: instead of presenting a traditional result list broken down by the availability within each service catalogue, each list is organized by content type (film, series, show, game, music) factorized, allowing the user to read once and choose where and how to enjoy the content.
Display the very same results on all platforms, striving for coherence in terms of content, yet adapting the interaction to the device. Adapting recommandation algorithms for « absolute 0 content »
This new search experience is being currently deployed. Continuous testing on prototypes involving users and stakeholders throughout the project allowed us to estimate the perceived speed and ease of use through the AXE methodology in the beginning. This is the single most effective step of the project in terms of internal cooperation. It allowed for a shared design vocabulary and understanding of the project. Furthermore it rapidly helped us gain major stakeholder championing.
As the meta engine reached full development, we then introduced more physical measures in pertinence and overall performance gains. As for internal cooperation, it is still in the early stages and difficult to assess how sustainable this evolution will be. Yet there are encouraging signs that this method is starting to spread on other projects within the company.