CarMax Onboarding Flow

Scroll slowly...

A crisp wind cuts through Kalhorne's Forest.

"Three days. We've been walking towards the Tower for three days and are no closer than when we first started."

You know the seer is right but...

Before you can utter a single word, silence falls upon the forest....

The sound of cracking wood simmers in the distance but soon it grows louder and louder. You start to realize the trees themselves are moving and reveal an incredible sight.

Sunlight cuts through the trees of the forest and in front of you lay two pathways. Your instincts tell you that only one of them will lead to the Tower, but which do you choose?

To take the left path, turn to page 31.
To take the right path, turn to page 44.
To turn around, turn to page 58.

CarMax Onboarding Flow

2.5 min. read

Product Designer:

Myself

Product Manager:

Kartik Swamy

Delivery Manager:

Michele Eckert

iOS Developers:

Jonathan King / Tyler Garrett / Raj Bedi

Android Developers:

Josh Sennett / Sayed Rohani

‘Choose your own adventure' stories were a staple of my childhood. Reading the Rumpelstiltskin book over and over again while trying to avoid the bad endings has become a core memory. But the possibility of getting a new story by changing just 1 decision is what kept me coming back.

In 2020, I used my childhood experience as inspiration when creating CarMax's onboarding experience. Design, research, and storytelling came together in a beautiful blend which led to my most successful feature.

Chapter 1: Call to Adventure

Let the book puns begin...

What if we didn't have to guess what our first time users wanted to do on the app? What if we created the best first-time experience because we knew what they wanted they even download the app?

An entire quarter was dedicated to answering these two questions. Using our analytics team, we took the most commonly used search filters, converted them to tasks, and used those to build our first ever ‘choose your own adventure' user test.

Figure 1. This is the overall structure of our user test. Drop them in at the top,
let them make their choices, and then document their final AB combination.

Chapter 2: The Road of Trials

The test itself was a series of questions posed to the user one at a time. We gave them a prompt, asked them a question, and then gave them two options to choose from.

Each step of the way, we either questioned their decisions or asked them to clarify their reasoning. This provided us with a look into their mindset while they traversed our test.

Pretend you downloaded a car shopping app for the first time. What are you looking for in your next car?

When you're done, select what you'd do next.

Create an account

( 9% )

Shop all cars

( 91% )

You've decided to shop all cars. What if I told you in order to save your favorite cars, you have to create an account?

When you're done, select what you'd do next.

Create an account

( 55% )

Select car type

( 45% )

You've decided to create an account. Please explain why saving your favorite cars made you change your mind.

When you're done, select what you'd do next.

Select car type

( 96% )

Input ZIP or location

( 4% )

Figure 2. These are some of the questions used in the test.
The percentages indicate the amount of participants that selected that option.

Figure 3. By the end of our research phase, 55 participants chose their own adventure.
The thicker the line, the more participants went down that route.

Shop all cars

Create an account

Select car type

Set maximum price

Select car type

Input ZIP / location

Set maximum price

Input ZIP / location

Set maximum price

Select car features

Input ZIP / location

Select car features

Input ZIP / location

Enable notifications

Select car features

Enable notifications

Chapter 3: Apotheosis

Most users wanted to go right into shopping for cars and did not want to create an account. But over half of the users would create an account if they knew they could not save cars without one.

Boom. Now we have a value prop to start the flow.

“Make an account to save
your favorite cars.”

Chapter 4: The Ultimate Boon

With our test complete and value prop in hand, it was time to design, iterate, design again, and iterate. Over and over again. We analyzed the results as much as we could and listened to the recordings to take away as many learnings as possible.

This process was the longest and most thorough but provided the biggest benefit to our team. We collaborated together and built a testing plan we were all proud of.

Chapter 5: Master of the Two Worlds

By taking a real world experience and applying that methodology into a digital world we created a highly successful first-time experience for our app users. Each screen in our new onboarding was carefully considered to achieve a specific goal. Each screen led the user down the funnel towards buying a car. Each screen was made possible by taking a childhood experience and creating a beautifully crafted story.

Shop all cars

Create an account

Showcase the value prop but give them the option to continue without an account.

Select car type

For the most part, customers knew the type of car they wanted.

Set maximum price

Input ZIP / location

In order to create an accurate budget we ask the user to input their ZIP code.

Select car features

Fun fact: Customers are more flexible on final price than they are on the features inside their cars.

Enable notifications

Using all of their answers, we give them a personalized 1st time search experience.

Final words...

Mission Complete 🫡

850

Increase in YoY sales from users
that completed onboarding

153

Increase in account creation
for 1st time visitors

2.2

Increase in repeat visits from
onboarded users

Epilogue: Freedom to Live

Minimum viability and great design don't always intersect. For a feature that was a personal journey into research methodology, I wasn't satisfied with my initial design. The second iteration, which never launched, is here for your enjoyment.