Predictive UX

Brand
Mercedes-Benz

Year
2013

Launch
DEC 2014 —
CES, Las Vegas, NV, USA

Role
UX Designer

Contribution
–> Creative Direction
–> Interaction Design
––> Visual Design

Recognition
Shortlist IXDA Interaction Awards 2015

Design a practical and understandable way to demonstrate how machine learning can improve our overall vehicle user experience.

Experience Concept
Driver distraction is a real issue. In order to simplify and reduce interaction steps with the telematics system, Mercedes-Benz R&D North America developed a prediction engine.

While using the system in the car it learns the habits and preferences of the driver. After a few weeks the system starts providing meaningful predictions to the driver. These predictions are presented as shortcuts into the system on the home screen of the head-unit.

Through these shortcuts the driver can start route guidance, initiate a phone call or switch to a different media source with just one single tap.

Interaction Design
To present this innovation at the CES 2014 to the public in a practical and understandable way we developed an interactive table in collaboration with our machine learning engineers and IconMobile.

The table can be used to create an environment for a virtual person that is driving a Mercedes-Benz. Based on this created environment the software calculates new predictions in real time. The thinking process of the machine was visualized on an additional screen above the user.

Visual Design
We used the visual design that we developed for the Concept
S-Class Coupé just a few month ago – both for the head-unit as well as the additional screens.

Design Process
—
Discover and Define

Understanding the User

Richard from Los Angeles, CA

"Enjoy Life While I Can"

Lifestyle

How and where does Richard live?
What kind of other brands does he value?
And how does he spend his spare time?

Personal Car History

This guy really is into cars, he owns and
has owned a lot of different iconic cars.
He knows everything about them.

Thoughts on Comptitors

Understands the landscape and
has strong opinions.

Nothing but the Best

Richard is a huge fan of the S-Class,
but expects the best for his money and thinks
there is room for improvement.

To Buy or Not To Buy

There's a lot of attraction but also
things that are lacking.

Core User Needs

UX Framwork

Technology should be used to understant the user in order to provide an excellent experience

Benchmark
—
Everyday Anticipatory UX

Predicitive typing and
auto-complete

Siri predicitions in Spotlight search

Search engine auto-complete and predictions / suggestions

Predicted and suggestes destinations in Apple Maps

Instant commute and traffic prediction in Waze

Siri watch face — UX designed around constant anticipated and suggested actions

Principles
—
Anticipatory UX

support the user

offer information or actions when and where needed to simplify the interaction steps to successfully complete a task

touch of magic

by integrating predictions and intelligence deeply into the core system, we can achieve a simple, positively surprising and fun to interact with user interface, that adds a certain touch of magic to
the system 

affective experiences

by (seemingly) understanding the user’s actions and intent, as well
as building trust in the car’s abilities

make it easy

using any intelligent features should not show the complexity behind the technology, but rather offer the user an easier way to complete tasks

——> suggestions as shortcuts to frequently performed actions

——> automatic adjustments for
invisible support

make it discoverable

for suggestions to be useful, users should be able to stumble upon them while using they system — suggestions hidden in a menu don’t add value and defeat the purpose

——> in-app “stumble upon” suggestions

——> contextual infos and actions as interactive  notifications

make it mine

While supporting users proactively in achieving their goals they should feel in control at any time and have the ability to make decisions for themselves

——> settings for data tracking

——> trust building interfaces for
autonomous driving

Make it Easy

direct action

Suggestions serve as shortcuts to actions
a user performs frequently.

The user doesn’t have to go through numerous
steps and menus to perform the action.

consistency

Use of inconsistent patterns can
lead to usability issues.

Discoverability

stumble upon

Suggestions should be located in a frequently visited and prominent location

proactive

Alternatively suggestions should be presented
to the user proactively.

Make it Mine

control

At any point, users should be able to turn off tracking or suggestions altogether

more control

Options are available when needed,
but are not first priority

Support the User

silent helper

Solution to undo possibly unwanted options could utilize
the voice system or push notifications.

multi-modal interaction

Solution to undo possibly unwanted options could utilize
the voice system or push notifications.

Companion App

Some suggestions can be activated automatically to simplify the user’s interaction with the system to a minimum

Connected Home

Adjust the temperature at home — 
just in time for my arrival

Anticipatory AC

Switch the AC in the car to circulate
the air, rather than taking in air from the outside,
when driving into a tunnel