Women’s Football: Using Voice Assistants to Educate and Inspire — PART II


Continued from an earlier blog post: ‘Women’s Football: Using Voice Assistants to Educate and Inspire — PART I’

Exploring the Process of Designing for Voice Assistants

Designing a VUI (voice user interface) for voice assistants radically differs from designing for a more traditional user interface, that is accessible through a screen, as there is an extra layer of complexity to consider in terms of how a person may interact with it. There is very little control over how a user interacts with a VUI, especially when thinking about it from a conversational perspective. Any question can be asked at any point, and the conversation can be immediately shifted to something entirely different in an instant.


When interacting with a website or app, a user begins the flow that has been considered and constructed by the designer. There is always a specific entry point in which the user starts to begin the flow, whether that’s starting at the home page, onboarding screen, subpage etc. A designer would always have an idea of what a user’s path is going to be, allowing them to consider the user’s point of entry and how they may be guided through a specific path. There is always a specific number of options they are able to interact with or can input at any given point of the flow, they can also immediately navigate to a completely separate area of the website or app, based on the options available or set out from the designer.


The structure of a user flow for a VUI is quite different. The user controls what their entry point is when they engage with the VUI and often this could be something that hasn’t even been considered by the designer. Plus there is no control over how the user will interact after a response is given from the voice assistant, or how their tone, clarity and volume of speech may affect the engagement with the voice assistant—these are a number of different areas to consider when navigating the VUI as well as when dealing with error handling.


User Personas

The first stage of the process was to define the people that will be engaging with this VUI. This is the basis to all User Experience led projects and because this was using a natural human experience it was key to understand the users entirely. The demographic I chose to focus on were children, around the age of 8 – 12 years old, predominantly female, with an interest or curiosity in football.


Each persona was formed based on real-life examples of children and their connection to the sport, derived from recent media coverage and numerous news articles that covered the uprise in the popularity of the female sport as well as the negativity and stereotypical attitude that also surround it.


User Persona for a young female child


As it was the first time creating User Personas for children, I wanted to compare our current process of constructing User Personas and use it as a basis to evolve the format. User Personas are all different depending on the user it is based on, and not all people will fit into a pre-constructed template. For example, a user’s occupation and salary won’t apply to a child.


A fantastic resource that helped to evolve our User Personas was this article from Beaker & Flint, it focuses on the fact that each template should be different and the User research should inform the layout, rather than letting the layout constraint the research.


User Journey and Empathy Maps

User Journeys are usually used to understand a user’s pain points and positives within a current product. As this didn’t really apply within these circumstances I instead wanted to create the user’s current journey to form an understanding of where the voice assistant could be used to help form a better experience for the user.


User Journey for a young female child


This helped to figure out at which points a voice assistant would benefit the user and the scenarios a user might be in when using the voice assistant. As well as understanding a user’s journey, it’s key to also understanding a user’s feelings.


Once the users have been formed and expanded on, and the journey they may take when using the current or proposed product has been finalised, an empathy map can be created in order to understand a user’s feelings, allowing a level of empathy to be formed with the user.


The more detailed a User Persona is the better foundation it creates to base your product on. The more you understand about each user, the fewer assumptions you have to make about how to design a product for them.


Empathy map for a young female child


Scenario Mapping

After identifying users, their needs and frustrations, I began mapping out a relevant scenario in which they will be involved when using the voice assistant.


Voice Assistant devices, especially Amazon Alexa and Google Assistant, are often stationary devices set in a specific room or area of the house and the environment in which it’s used will often change based on the time of day. There are a number of categories to factor in when creating a scenario map, for example, who will be using it, the type of person they are or the mood they may be in when interacting with the voice assistant, the brand or character of the voice assistant as well as a variety of other examples:


Who: Someone (or a group of people) using a VUI.

Social context: Are they at home? At school? In a car?

Emotional state: Are they inquisitive? Tired? Task-focussed?

Brand / Character: Who is the user talking to? What kind of content?

The tone of voice: Peppy? Sensitive? Teacher-like?

Device(s): Amazon Echo? An in-car system? A smart radio?


Once all of the above had been established I began to consider the questions that could be asked by each person using the VUI:

  • Could you give me an overview of [What is] the FIFA Women’s World Cup?
  • What year did the FIFA Women’s World Cup start?
  • How long does the competition last?
  • When is the World Cup Final?
  • Where is the competition held?
  • Who won the last World Cup?
  • Who has won the most World Cups?
  • How many teams compete?
  • How many groups are there?
  • What teams are in {e.g. Group A}?
  • Who is top of {e.g. Group B}?
  • Who is bottom of {e.g. Group C}?
  • What are the chances of {e.g. England} winning the World Cup?
  • Who is playing today?
  • Who is playing this week?
  • What games were played yesterday?
  • What games were played last week?
  • What were yesterday’s results?
  • What were last week’s results?
  • Who’s playing next week?
  • Who got the assist for {e.g. England}?
  • Which player assisted in the {e.g. England} v {e.g. Argentina} game?
  • What player received bonus points in the {e.g. Southampton} game?
  • Which players scored in the {e.g. Japan} versus {e.g. Scotland} match?
  • Who got a card in the {e.g. South Africa} game?
  • Who has scored for {e.g. England}?
  • Who scored in the {e.g. France} v {e.g. South Korea} game?
  • What’s the highest score this week?
  • Which players play for {e.g. England}?
  • Who is the {e.g. England} captain?
  • Who has the most caps for {e.g. England}?
  • What are the next fixtures for {e.g. Brazil}?
  • Which opponent do {e.g. Japan} face this week?


This bank of questions would then be expanded on as various user tests are carried out. These questions are based on an assumption of what I think the user may ask but may not always be the case. By conducting a role-playing session, in the form of a post-it note script I was able to gather a clearer understanding of how a user may interact with this experience.


Post-it Note Scripts and User Test Role Playing

When defining a user flow for a digital product such as a website or app, you have a greater understanding of which possible route a user may take as they are limited by the number of options presented on screen. When designing a VUI you have very little control over how a user will interact with the voice assistant, there will always be a user flow in place however it won’t be as apparent to the user as a website may be. On the other hand, a user can quickly find out what they are looking for by asking a question.


Figuring out the type of questions a user may ask and the interaction they may have when engaging in conversation with the voice assistant can be tricky. I originally created a bank of questions in which I predicted what a user may ask or inquire about when using the skill, as well as an introduction and a starting point that allows for the user to be able to engage with the voice assistant. For example, once the user has opened the skill Alexa would introduce herself and then ask “What would you like to know about the FIFA Women’s World Cup?”. This then initiates the conversation and allows the user to answer the question in response.


Whiteboard session to figure out flow for Voice Assistants


As this response could completely differ from user to user, creating a rough script using post-it notes allows you to imitate Alexa and her response while figuring out what a user may ask. It also allows for quick iteration and position of various interactions throughout the process. This is especially relevant when creating a VUI that will engage in conversation as opposed to answering a single query.


Having a set of post-it notes that contains each response that is available allows you to build a wide variety of conversation strings. These strings are also able to be connected to one another depending on what is asked and where the conversation leads to. The best process I found was to add a post-it note with the user’s response in one colour and then add another in a different colour for Alexa’s response. This then begins to form a user flow between both the voice assistant and user and allows you to consider where to go next throughout the interaction. BBC UX&D followed a similar format when designing a VUI concept for children:


“We also add notes for actions into the flow – for instance, a machine looking up some data, or a person adding ingredients to a recipe. These actions are also colour-coded by the actor. You could also add notes for multimodal events – things like information appearing on the screens of nearby devices, or media being played on a TV.


Sketching out a voice-led experience like this is a really quick, useful way to see how its flow and onboarding will work, and which bits seem uneven or need more unpicking. Using sticky notes means that it’s really easy to add and remove events or re-order sections where needed. Using colour-coded sticky notes means you can easily see if there are parts of a flow where the machine is saying too much, or asking too many questions.” – BBC


This workshop reflects a test scenario in which you can have a user question you as if they were talking to the voice assistant. This immediately gives a feel as to what works and what doesn’t, how a conversation may develop, as well as ways to get back to the original topic so that other areas can be explored.


Post-it note session to figure out flow for Voice Assistants


In this instance, having a conversation about a specific event and its history can span off onto a number of different conversational areas whether it’s about the success of certain teams, an overview of the players, a stadium’s capacity, or the current and past standings or results. It’s especially relevant so that conversational loops can be avoided. Having a user be stuck in a conversation loop without being able to get out of it is a point in which a user will immediately drop off and will be less likely to engage with the voice assistant.


“If you can, try to recruit someone to act as the user who isn’t already familiar with the idea – they should be ‘using’ your prototype in as real a way as possible. This will enable you to get early feedback and observations on how the experience flows and any pressure points or parts where the ‘user’ feels lost or unclear about what’s happening.” – BBC


Once any issues have been resolved after working through a number of role-playing scenarios, a final number of user experience scripts can be written to demonstrate user dialogue flows in order to finalise conversational strings.


If you would like to know more about VUI’s (Voice User Interfaces) and how they can be used within your organisation we’d be happy to open a conversation with you to see if we can assist.

Drop us a line


Prototyping VUI Conversations with Dialog Flows

By recreating the post-it note scripts in a digital dialogue flow map you can begin to finalise what keywords, or trigger words, can lead to what interactions. This then gives a clear image as to when the conversation may branch off and where it would lead to, as well as how the user may get back to a specific point of dialogue.


“A dialogue flow is a script that illustrates the back-and-forth conversation between the user and the voice assistant. A dialogue flow is like a prototype, and it can be depicted as an illustration, or there are prototyping apps that can be used to create dialogue flows.” — UX Planet


Keeping both the dialogue boxes for both the user and voice assistant in strongly contrasting colours allows for an easier understanding of both points of engagement. There are some points at which the response can be variable in regards to date and time, for example, group stages, match results, standings etc.


Dialogue flow to figure out flow for Voice Assistants


The complexity and scale of a dialogue map could be hugely vast, due to the sheer amount of inputs from the user, conversation strings, a variety of answers and tangents in which conversations could go. It is almost impossible to map out every instance of the dialogue map fully. The importance of user testing and carrying out role-playing scenarios as much as possible is key to making sure that you have considered every instance of interaction between the user and voice assistant, rather than assuming how a user will interact with the VUI. Each user will interact with a voice assistant differently and testing this extensively is the only way to consider every outcome.


Read: ‘Women’s Football: Using Voice Assistants to Educate and Inspire — PART III’


Valuable Resources and References



Curiosity is what drives us to learn and grow professionally enabling us to remain vigilant to change and the opportunities it presents for our clients. We document some of this learning and share it via our 'Insights' section.