Radio France International
Ma RFI App Concept
While studying for the UX design certification at the Gobelins Design School, we were tasked to work in teams with a real life customer on a long term project. Our challenge was to design a new and innovative experience around information and radio in 2025 for France Media Monde, one of the leading broadcasting companies in France. We focused on RFI (Radio France International) which is a public radio broadcasting its programs in French all over the world, but mostly in Africa. Their goal is to conquer a new digital audience while staying true to their values. We therefore imagined and designed the Ma RFI smartphone app, which embodies a switch from one way communication to a real conversational radio, fitting their current and future users. I was involved in all stages of the project but was more specifically in charge of user research and testing.
Understanding the customer
Before going into the actual user research, we took the time to better understand our customer and his context. This was a critical success factor for the project as it allowed us to better identify the expectations and challenges surrounding the project.
We first of all realized at a kick-off with the different stakeholders that there was no clear problematic defined and that our first task was going to make them converge towards a single design problem to solve. We soon also figured out that there were other important aspects which we would have to take into account. FMM is a public company and thus financed by the French government which had quite a few political implications. At the same time the company was going through a digital transformation process, which meant that clear objectives had been set in order to work in a more agile way and develop a more innovtive offering.
RFI's current digital ecosystem was also very fragmented, resulting partly from a complex and siloed organization. Mapping all the different websites, apps and interfaces was a challenge in itself. Last but not least, we conducted a competitive analysis in order to better assess the strengths and weaknesses of RFI's direct (CNN, BBC, Al Jazeera...) and indirect (newsfeeds, social networks, television...) competitors.
Secondary research
In this phase our goal was to understand how people were staying informed today and they eventually would in 2025, based on the current prospective technological trends for the radio of tomorrow (vocal assistants, personnalization...). Additionally to the fact that usage is getting mobile/smartphone focused, even in Africa, these inspired us in shaping our solution. Even more importantly, it was key for us to define what "information" actually is and all the shapes and forms it can take, as we needed to figure out the expectations and context for each of them for our user research.
User research
We used several research methods in order to have a better knowledge of the current and future RFI radio/digital audience, and ultimately our end users. We needed to better understand what their relationship with information in general was, how they were staying informed, their habits, how their day looked like... This helped us identify their needs, motivations and constraints in doing so. The synthesis and output of this research was the creation of user personas and the formulation of different hypotheses for our potential solution.
​
Quantitative research: We were thankfully able to access a lot of existing documentation and market research from RFI, but also to send out a survey to a panel of users through their website, to have even more relevant quantitative data available. Some key insights were generated such as the fact that they were a majority of smartphone users going on the internet several times a day or were mostly interested in international information. Their main concern for 2025 is that they wanted quality information that they could trust.
​
Qualitative research: We conducted 28 interviews to have a more qualitative view of things. The interviewees were from all ages but half of them were from French speaking Africa as this is were a big majority of RFI's audience is located. We made also made sure to have both RFI listeners and non listeners. The goal was to discuss their activities and attitudes towards information as well as challenge them on their ideal vision for the future of information in 2025. This phase was probably the richest in learnings. Again, trust and neutrality were identified as key factors for RFI which has a very loyal audience.
Personas: Before modeling our personas, we had to define behavioral variables on which we could place our different interviewed users based on the data collected. These were based on the actions they were undertaking in order to get information and their preferences in terms of information type. They enabled us to identify behavioral patterns and have different archetypes emerge from those. We soon realized that our customer had high expectations on these personas and we had to do quite a bit of UX evangelization so that they understood that they were design tools and not deliverables. We therefore deliberately removed all sociodemographical characteristics or names/pictures from them so that they would not associate them with their marketing personas. 3 different profiles were surfaced in the end
The Demanding
They like to be at the heart and source of information while having a true and deep understanding of it
The Insatiable
They needs to be first ones to be informed and to be constantly updated, but don't want to spend too much time doing it
The Epicurian
They let news come to them and need it in an easily digestable and playful way
For each of these of these personas we have mapped their user journey throughout their day (with actions, thoughts, feelings, and experiences at each stage) to identify opportunities for our potential solution in the problems it could solve. These user maps were very useful in the design stage as we could storyboard different usage scenarios for our app.
Redefining the design problem
Our original scope was very large and we needed to focus on a more specific problem to solve before going into the ideation phase. We built on all our previous research, the identified potential solutions, the constraints and the wished/expected impacts for RFI as well as for their users. We therefore converged together with our customer and worded it as such:
How can we conquer a new digital audience in 2025 for RFI while developing the link between diasporas in France and their country of origin?
We also decided at this stage to define 4 design principles which would give an more precise orientation for our solution and remove any ambiguity:
​
Sound above all: RFI is first and foremost a radio broadcaster and voice will be at the heart of tomorrow.
​
Exchange: RFI also acts as link between communities. Sharing, exchange and interactivity shall be key.
Personalization: RFI is a public media for all and personalization helps cover the needs of each and everyone
RFI spirit: The solution will need to stay true to RFI's values of trust and neutrality which its audience is strongly attached to
Ideation phase
Before we could reach a high fidelity prototype we went through different stages and ideation workshops in order to converge towards a single solution:
​
- User map workshop in order to understand our users constraints in terms of access to mobile data during their day, as this is a key issue in Africa​
​
- Crazy 6 workshop which helped identify opportunities and constraints in terms of personalization
- Low-fi prototype which we guerilla-tested in the street, which gave us a lot of new ideas and potential improvements
​
- Last Crazy 8 workshop before wireframing, building on the ideas and inspiration gathered in the previous stages to define more precisely the final product
- We then designed a user flow, sketched it on paper, wireframed it and implemented the UI components in Figma for the prototype to be tested
Ma RFI
The result of all this hard work was the "Ma RFI" smartphone app, as mobility is key dimension for our current and future users.
They can interact with the app through a vocal assistant or a chatbot. It has deep learning functionalities as it bases recommendations on previously read or saved elements.
​
Another central feature is the concept of "RFI Ambassadors" which are local non professional correspondents vetted by RFI and can be a relay for the hottest local news for the community, helping our users keep a link with their country. Their content can easily be differentiated from "official" RFI through clear colour coding.
We also designed a fast and intuitive onboarding and personalization process around 4 questions to answer which will impact the users personalized news timeline. Nevertheless users can easily switch back to a non personalized version of the app so that they can be sure they don't miss out on anything, from current shows to live broadcast and bookmarks.
We decided to completely design an interactive prototype of the "Explorer" module (which we later tested with real users) where users can view the latest news on a map through country/city choice or geolocalisation. These can be filtered according to preferences or viewed as a list.
​
The personalised timeline adapts itself automatically according to the usage context. For example when driving, podcasts are automatically programmed according to the planned trip duration, or articles can be read out to the user if he's busy.
User testing
These tests allowed us to evaluate and confront them to our potential users. Combined with qualitative interviews we were also able to (in)validate some hypotheses, such as whether users could understand the concept of RFI Ambassadors and if and how they would like to interact with the app by voice. 5 users of various ages participated and were recruited in our school. They were all "digitally mature" but were not podcast listeners or users of vocal assistants even though they were used to reading news on a mobile device, which was a good opportunity to see how our concepts would be received by a new audience.
​
The tests took place in the Gobelins School's ErgoDesign test lab and were organized following these steps:
​
-
A pre-test questionnaire in order to identify the profile of our testers, their habits in terms of information and digital practices.
-
The actual test which was conducted on a smartphone, which was recorded with a webcam and a microphone. We used an oculometric device in order to record what the testers were looking at on the phone's screen. They had to follow specific instuctions and verbally comment what they were doing while being observed by two of us via a difficulty matrix.
-
A post-test questionnaire in order to better evaluate how easy it was for them to follow the instructions and how they perceived the interface (though a User Experience Questionnaire based on pre-defined conditions). This was followed by an interview so that we could dig deeper into the testers feedbacks and show them the rest of the already designed screens.
We were very aware that these tests had a lot of bias and that we had tested only small part of the app (focusing on the "Explorer" mode) but it help us take a step back from our work and introduce some changes. Overall the feedback was positive but there were some recurring issues (such as on finding a specific article) and the concept of Ambassadors was not always properly understood, which led us to rename them as "Scouts". We also made several items and typos larger in order to increase readability and usability.
Conclusion
This excercise was very particular because of its prospective nature and that the final product was not made to be developed in the very near future, even though we tried to find the right balance between innovation and RFI's constraints. "Ma RFI" will mostly serve as an inspiration for future projects and should ideally be implemented gradually in a continuous improvement approach according to the company's agile framework. On a more personal level, this project allowed me to explore a field which is very different from the retail e-commerce customers I usually work with and apply several newly learned skills on a complete UX project from A to Z.