Volkswagen – Capture App

A design sprint about the question, how can we enable commuters to be productive while driving by car.

647a1565ce3735a3418e1732_texting and driving-min-1

Role and team

Sprint Master (Microsoft), Product Owner (Volkswagen), UX Designer (Me), Developer (Microsoft), Research & Test Users (Volkswagen & Microsoft)

Responsibilities

Reseach, UX Design, Conversational Design, Prototyping

Timeline

10 days

Deliveries

Personas, Feature Map, User Scenarios, Task Analyze, User Flows, Graphical Interface Design, Voice Interface Design, Prototype, User Test

Challenge

In this design sprint we used the Product Thinking method to solve a specific problem for a specific user group through a product or service offering.

How can we enable commuters to be productive while driving by car?

Design Goals

When we planned the design sprint on the first day, we also agreed on the most important evaluation criteria for UX.

Efficiency: Ensure that the time needed for users to complete tasks is short and the users focus stays on the productivity aspect, now the interface itself.
Simplicity: Ensure that the user interface is intuitive and easy to use through simplicity and clarity.
Satisfaction: Ensure that tasks, that were started while driving, can seamlessly be completed afterwards and that they make the user feel satisfied (happy) through clear user feedback.

Design Process

1. Discover

USER PROBLEM – What problem do we solve?

In the discover phase, a discussion began about what the problem is and how the problem feels like, how big the problem is and whether a solution to the problem would mean real added value for the users.

To find out, we conducted a survey of internal employees who we knew commuted. We extracted the desired features, connected services and the main problem statements from the survey:

I cannot work (office work) while i commute by car.
I cannot organize my work while i commute by car.
I cannot capture moments of inspiration while i commute by car.
I cannot capture ideas and to do`s on my commute by car.

TARGET AUDIENCE – Who is it for?

While discussing who the target group is and it was already clear, that it ithe group of car commuters, we identified further segments within that group:

Commuters by car
Commuters by car who travel the same route every day
Commuters by car to different destinations
Commuters by car with short journeys
Commuters by car who travel very long distances

2. Define

VISION – Why are we doing this?

The final product vision would be realizable through the technical development in another area, the autonomous driving with cars. This would ensure full productivity during the journey. The product would therefore be an interim solution on the way there and would generate valuable data in the process.

STRATEGY – How are we doing this?

Knowing that this is an interim solution and in order to establish a strategic cooperation with Microsoft, on whose cognitive services the initial voice interface is based, the service is to be offered to a broad audience and not just to Volkswagen customers. The service is therefore being developed as a smartphone app and is compatible with all types of car in order to be able to collect as much data as possible.

GOAL – What do we want to achieve?

Knowledge in the aera of In-Car Productivity and the use of voice user interfaces while driving.

FEATURES – What are we doing?

We created a Feature map that was already structured in epics, features and functions to define the projects scope and as a first step towards user flows.

3. Bulid & Test

We sketched out the user experience in the four user scenarios, based on the four main problem statements and did a Task Analyze of the ideated user flow. We then created wireframes and dialogs of the corresponding User Flows as preparation of the Prototype. The Graphical User Interface was created with Axure and we simulated the Voice User Interface with pre-recorded phrases, spoken by Mircrosoft’s Cortana.

Sprint Planning

A design sprint unfolds its magic through the dynamics that arise over the reduced period of time. In order not to become isolated on individual tasks, strict time planning is relevant for the success of the sprint.

What`s the problem?

While commuters on the train can use their time productively on the way to work, drivers are obliged to concentrate on the road. The legislator has specified the maximum duration in which one can keep his eyes off the road while driving, but it is clear that the visual and mental focus should to stay on the traffic situation and perspectively decreases with the use of further safety technologies.
In contrast, on the train one has the disadvantage that the entire compartment can listen to ones video conference or telephone call. Using a Voice Interface in a train might also lead to more knowledge about personal goals and activities, than one would like while the usage of Graphical Interfaces are not a problem at all.

For whom are we doing this?

Strategy

Aware that there are devices that are basically constantly listening and waiting for a keyword, such as Alexa, it was decided in this project that users should have full control over the listening mode – by actively starting and stopping it.

The service should also be compatible with a wider user group, not just Volkswagen customers, so the plan is to use third-party Bluetooth buttons to start and stop the voice user interface, regardless of what type of car is driven.

Features

The MVP scope and the features were extracted from the target group survey, defined by the product owner and collected in a Feature Map, that is already structured in epics, features and functions.

Build and Test

We outlined the user scenarios according to the four main solved user problems in flows. After sketching and letting the imagine run wild, the scenarios were cleared and analyzed in User Goals as Use Cases:

The Use Cases were:

Login to a private or business service account.
Capturing an idea and forwarding it to the own email adress.
Creating and changing appointments and sending it to the own calendar.
Creating and structuring ideas and a list and sending it to an other online service.
Capture an inspiration for a blog (a long text) and keeping it as a transcript for later. Beeing remembered by the service after 7 days.

User Scenario Sketches

Task Analyze

Dialog Design

The dialogs were created under the precondition, that the congnitive services can not only transcript audio into text, but also „understand“ the users intention so that the user can give clear and simple commands it also would react to predefined keywords, like service names that the user e.g. want`s to send a transcripted capture to.

Wireflow

The Graphical Interface is sketched with OmniGraffle as wireframes first and later designed and prototyped with Axure. This interface should not be used while driving but also offers fthe ull functionality of the service while not driving.

Screen Design

User Test

Because of the limited time, it wasn`t possible to implement logic to the voice user interface. We created core dialogs and recorded the spoken sentences by MS Cortana as audio files, which we played back from a tablet during the user test and simulated the Voice User Interface that way.

The entire test was recorded and evaluated with a GoPro camera. The Graphical User Interface was tested via a clickable Axure prototype. The perspective users were asked to compete concrete tasks and interviewed after the drive.

Conclusion

1. A try-and-error mode while using the Voice UI

Not everybody was used to Voice Interfaces back in 2016 and the human-machine-dialogues often went into a trial and error mode, looking for commands that would be understood or accepted. This could also result from the test setup, but what we can say is, that nearly everybody was trying to explore the boundaries and the cognitive capabilities of the service. So a initial voice menue might have taken the users more time to finish the tasks, but would have given more guidance and a better overview of what is possible.

2. A seamless service would be prefered

The users did like the features of the service, but the design goal to create a user flow that feels satisfying when ending a tasks could not be fully achieved. The users didn`t like the usage break that is generated when continuing a task that started with capturing via the Voive user interface and had to be finished with the Grafical user interface after finishing the drive.

3. A more flexible differentiation between private and business usage

Even though the users demanded a clear separation between private and business content, the offered solution to clearly devide the content when logging in to a private or business account was seen as too unflexible, because it prevents users to capture private and business ideas on the same route. An initial neutral recording of ideas and specification based on the recipient would have been preferred.

Learnings

This was one of my first design sprints in 2016 and because there were no iterations planned, the main result after user testing was the evidence that emerged.

It was interesting to see that evidence does not always come from the effort or time put into individual tasks, but that in an ideal world, UX design is a dynamic cognitive process that impacts a group.