Porsche: A Service Design Case Study
Position: User Experience Design Student, UCLA Extension
Project Type: Service Design Project
Project Role: One of three UX Design collaborators
Timeline: Sept. 2020 - Dec. 2020
Users: Porsche Employees & Porsche Clients
Project Background
This case study is the product of an 11-week service design intensive. I was paired with two other students, from the UCLA Extension UX Design Certification program, to collaborate on a solution for a service design challenge.
The Design Challenge: How might we design a holistic solution for luxury car dealerships operating during the COVID-19 pandemic?
Project Summary
We applied design thinking to re-adapt the auto dealership model to the current environment. Ultimately, our solution, called the “Porsche Pavilion”, reimagined each and every touchpoint within the dealership experience: from browsing to purchasing to servicing.
We leveraged emerging technologies and a deep understanding of Porsche culture/buyers to quell COVID-based concerns and make the purchasing and service experience simpler, safer, and more enjoyable.
The resulting feedback from buyers surveyed stated that they felt safe from COVID at the reimagined Porsche Pavilion (5 out of 5 on the Likert Scale). Surveyed individuals also believed that they had a unique and elevated experience at the Porsche Pavilion (10 out of 10, SUS Score).
The Problem
The luxury auto buyer expects an exciting, safe, and comprehensive end-to-end vehicle ownership experience, even during a worldwide pandemic. Given that all buyers have a different perception of the threat and varying tolerance levels for risk, dealerships and buyers alike do not always agree on what “safety” means. Because of this, dealerships adjust their policies reactively, resulting in operational strain that has cascading effects on the experience for both the workers and the client.
The Process
Discovery: Understanding the problem
Preparatory Research
To develop a complete understanding of the Porsche brand and the existing landscape it was operating within, we needed to define and prioritize our research. We built an assumptions grid to ensure that all efforts were moving us towards our end goal of defining the problem.
With our output from this exercise, we established our research goals, which would act as our north star during the formal research phase.
Listed out all of our assumptions about Porsche, clustered, and plotted them to get a sense of what our riskiest assumptions were.
Research Vehicles
Next, we prepared for our research phase by establishing the research vehicles we would use. This would guarantee high fidelity insights and avoid the misdirection of any efforts.
For instance, in order to understand Porsche buyers’ psychographics and demographics, we opted to carry out an online ethnography of existing Porsche communities and interview several Porsche owners.
An added benefit of carefully selecting our research vehicles is that we could triangulate our insights giving us greater confidence in our assertions down the line.
Remote Collaboration
Given the pandemic, our collaborative work was facilitated by Zoom, Slack, and other coworking tools. To bridge the distance, we developed specific workflows to simulate in-person collaboration. For instance, we carried out whiteboarding exercises in InVision. This gave us a shared workspace to collect our findings and carry out collaborative exercises for all to see.
Research Phase
The following account of our research process is the product of several cycles of team divergence and convergence. Any new assumptions or knowledge gaps that were identified at the end of a research cycle were acknowledged as a team and addressed during a new cycle of research and integration until we were confident in our findings and level of knowledge.
Desk Research: Our primary research vehicle, used to understand things like Porsche’s position in the market, its current adaptations to COVID-19, and its future plans for innovation.
Online Ethnography: Using carefully selected online sources, we’d learn about the Porsche culture. We looked at large volumes of self-reported customer feedback on Yelp. We observed dialogue and interaction between the Porsche buyer and the brand on Twitter. We gained an understanding of the existing communities on Reddit. We identified the niches, nuances, and general attitudes of the brand on Rennlist.
Interviews: We carried out covert interviews, cold calling 11 car dealerships to experience their service first hand, learning about their COVID policies as we went. We prepared interviews with 3 luxury car owners, in order to gain a deeper understanding of their experiences, emotions, and motivations. We carried out a contextual interview, with the General Manager of a Mercedes-Benz dealership, where asked about the specifics of their dealership and its touchpoints.
Definition: Stating the Problem
Research Wall: A consolidation of our entire body of research, the Research Wall served as a central point where our team could browse current research, list out new assumptions to tackle, identify research gaps, iterate our problem statement, inform our visualization models, and much more. Beyond housing important qualitative/quantitative data it served as the canvas for brainstorming/exercises, ideation initiatives, and prototyping efforts. Check out the sample screenshots below, or explore the entire wall here.
Personas: Our client personas would give us “characters” to base our future journey mapping and prototyping on. Given the sheer amount of ethnographic and psychographic data we gathered, our personas were rich in detail and we had a solid understanding of the Porsche buyer.
To see the full version of these personas, click here.
User Journeys: We used journey mapping to note key activities/actions, motivations, and thoughts/feelings that typically occur during each stage of the purchasing or servicing journey, all based on the detailed information we gathered in our interviews and ethnography. This made us aware of the major pain points and informed our future opportunities solutions.
The Solution: The Porsche Pavilion
We entirely reimagined the standard Porsche dealership model, calling it “The Porsche Pavilion”. This model would serve as an “ideal” for the future establishment of Porsche dealerships, as well as the retooling of existing ones. This solution defined an experiential template for how each individual touchpoint should look and feel, in order for a dealership to meet the changing needs of the Porsche shopper in the existing environment.
This dealership featured:
• A design with CDC guidelines in mind
• An outdoor canopy showroom allowing guests to shop for vehicles safely in an open-aired environment
• An automated personal shopping assistant, named Louise, for no-contact education
• Personal Protective Equipment for staff and visitors
• Solo test drives
• Open aired Garden Cafe
• Seamlessly-integrated luxury COVID protocol
Our finished proposal included 7 redefined touch points, with varying degrees of modification.
Example:
The first touchpoint, Valet Arrival, contained steps like reservation check-in, the gifting of safety swag, a temperature check, and informational televisions with digital signage to brief the guest on policy and expectations.
The touchpoint solution was not only limited to experiential design. Several touchpoints featured the use of a digital or audio interface.
• Pre-visit, online workflow for booking an appointment on the Porsche website
• An AI-powered, Voice UI assistant, that would be situated near each vehicle on the sales floor
Develop: Arriving at our Solution
With such an abundance of opportunities and pain points identified, picking the right solution to develop would be a challenge in itself.
Pre-Ideation
We took our Key Insights and identified the ones with the most friction or a job to be done. We’d use a tactic called “How Might We” to ask ourselves “How Might We remedy this particular pain point?”. If we observed the same question being asked multiple times, it would get clustered. These clusters would get prioritized based on impact and then ideated upon.
We wanted our ideation exercises to be open-ended and judgment-free, so we’d employ techniques such as Crazy 8’s (a collaborative sketching exercise meant for rapid ideation of outside the box ideas). We’d next pick and prioritize our favorite ideas using dot voting and cluster mapping.
Ideation
Ultimately, we landed on a rudimentary idea of the Porsche Pavilion concept. We started with a future-state storyboard to shed light on the buyer’s personal narrative throughout the visit to the dealership and see where we could innovate.
This is a small portion of the storyboard illustrating the Porsche buyer’s visit to the dealership.
This technique allowed us to approach the process experientially while inserting all of our COVID safety-related ideas from pre-ideation. The steps outlined in the storyboard would be either refined or removed over time, based on feedback gathered during iteration.
As the concept started taking shape, we’d naturally begin to approach this solution as a long string of interactions to be designed, each requiring their own unique solutions. Some required digital interfaces, others physical procedures, and some both.
Zooming In:
Sales Floor Touchpoint - Ideation
Let’s examine the ideation process for an individual touchpoint, to demonstrate the design thinking that went into it.
For the case of the Sales Floor, we designed a Voice-based digital shopping assistant named Louise.
Problem: Covid exposure and pushy salespeople are two of the biggest pain points we identified about the in-person shopping experience.
Solution: To mitigate these concerns, we designed a solution that allowed customers the freedom to shop without needing the accompaniment of a salesperson. Louise minimizes the COVID risk for both parties. Customers can always consult with Porsche employees if they want a more personable experience.
First iteration: A scannable QR code situated at every showroom vehicle, designed to provide general education about each vehicle. After some testing, we found this to be too limiting and technologically outdated. During subsequent research phases, we made a major pivot.
Final iteration: We created an AI-powered Voice Interface that possessed full knowledge of the Porsche vehicles on display at the dealership. This would enable a user-friendly, and contactless, way for car shoppers to ask any questions they may have.
Prototyping & Iteration
Given that the Porsche Pavilion was primarily a physical experience (as opposed to a digital one), we’d rely on a different type of prototyping to test our solution: the Desktop Walkthrough. Traditionally, a service designer doing a Desktop Walkthrough would create a small model of the buildings, obstacles, or objects involved in carrying out a service. They, then would use moveable characters to simulate the service, recruiting people to play roles and act out the entire process. Given that we were working remotely, we created a small model on a collaborative InVision whiteboard, with draggable characters.
Drawing detailed game boards with a mouse is…challenging.
Desktop Walkthrough Goals:
Understand all micro-interactions
Empathize with clients, map a dramatic arc
Identify areas of opportunity, reveal any knowledge gaps
Determining the feasibility of proposed solutions
Each one of these figurines played a role in our rehearsals. Each member of the team would play one or two roles and act out the procedures.
Workspace & Materials
InVision board, InVision “characters”
With each rehearsal, we’d extract new insights, update our documentation, do more research, and further iterate on our design.
Desktop Walkthrough rehearsals played a crucial role in the “Future State” mapping of every interaction. With this we’d understand its negative or positive impact on a clients overall emotional experience.
Zooming In:
Sales Floor Touchpoint - Prototype & Iteration
Let’s zoom back in on the sales floor touchpoint, in order to demonstrate “touchpoint” level prototyping & Iteration.
To prototype and test Louise, we built Voice User Interface flows for a variety of conversational scenarios. This allowed us to test her interactions with each other and improve our design.
Ultimately we prototyped and iterated on each individual touchpoint, and the entire experience overall.
Measurement & Success
To validate and measure success, we defined what success meant in our prototypes. Using various whiteboard and clustering exercises, we generated “success statements”. We then brainstormed the most effective methods for measuring each individual success statement.
For example:
Success Statement: ”COVID Concerns were Addressed”
Vehicle: Visitor Survey, usability testing for visit reservation prototype, employee-facing survey regarding COVID policies
From there we looked to establish a more quantitative standard for each success statement. Choosing from differing usability metrics like the Likert Scale, SUS Scores, or the Net Promoter Score, based on the success statement.
Testing & Measuring
We performed usability testing on 3 users with our Porsche Pavilion desktop walkthrough prototype. This included the entire experience: touchpoints with additional prototype builds like Louise and our pre-visit appointment COVID screening, and a survey at the end
Surveys
We built our surveys in TypeForm, creating one for each of the following experiences:
• General Client Pavilion Experience
• Louise Usability Survey
• Employee COVID Workplace Survey