Enhance new credit card provisioning with Walmart Pay

Walmart • 4 months (Jan – Apr 2020)

Overview

The new credit card is a co-branded product of Walmart and Capital One. The jointed partnership introduces an unprecedented, tightly integrated credit card rewards program. We wanted to give customers a streamlined shopping experience – so they can use the new card and earn rewards within minutes after approval. But our vision of a “magical” provisioning experience did not perform well post-launch. 

This case study is to walk through how we improved the provisioning experience to enable the new credit card with Walmart Pay.

Glossary

A few terms you’ll need to know:

Walmart Pay – a quick and secure mobile payment service within the Walmart app. It is exclusive for Walmart customers. People can check out with their smartphones in Walmart stores touch-free.

Capital One Walmart Rewards Card – The co-branded credit card launched on September 24, 2019. One of the card’s value propositions is to use it with Walmart Pay to earn 5% back in store for the first 12 months.

Text-to-Apply – Walmart customers can apply for the new credit card via SMS text message and mobile brower within a few minutes.

Provisioning – Enabling the credit card so it’s ready to use.

“Magic Link” – Upon the new credit card approval, Capital One sends Walmart customers a follow-up text message. It contains a link to add the new card as their default payment method in Walmart Pay. 

Typically, people need to wait for 7-10 business days to receive the physical card. With Magic Link, they can shop with the card in Walmart stores immediately. 

My role

I was the lead designer of the “Magic Link” provisioning feature, including the initial launch and post-launch enhancements, between April 2019 and April 2020.

The feature first launched with the credit card program in September 2019. The post-launch enhancement was designed January through April 2020.

I worked alongside 1 Design Manager, 1 Researcher, 1 Content Strategist, 1 Product Manager and 10+ Engineers. 

The goal

Walmart customers who apply and get approved for the Capital One Walmart Rewards Card via Text-to-Apply can easily add the card to Walmart Pay, use it immediately and start earning 5% back in stores.

• How we measure effectiveness:

• Increased percentage of customers who use Magic Link to add their new credit card to Walmart Pay (conversion rate);

• Increased number of customers who use the card with Walmart Pay (deeper engagements). 

The problem

Within the first 11 weeks of the new credit card launch, Walmart netted $2B in sales and Walmart Pay experienced two consecutive months of 18% growth.

However, our analytics showed 70% customers drop off the “Magic Link” provisioning flow from the Walmart Pay value prop screen and do not tap "Continue."

The "Magic Link" was not magical.

How might we…?

… make it clearer to customers that:

• Adding the new credit card to Walmart Pay is "magical," because we’ll do the heavy lifting and add the card to their account;

• All they need to do is start spending with the new card and earn 5% back on their current purchase;

• If they are not ready to use it now, we’ll encourage/remind them on their next visit.

It’s not as simple as just re-designing this one value prop screen.

The self-diagnosis began

Too many hoops to jump through

From card approval to use it, there were too many steps. We had been crossing our fingers that customers would jump through all the hoops that we set up. We delayed confirming “Card added” until we opened the Walmart Pay scanner. It appeared to be too late in the process. Most customers already dropped out before then.

One size does not fit all

Whether a customer has used Walmart Pay before or not, we showed them an identical value proposition. We also always open the Walmart Pay scanner to confirm, “Card added.” This was “by design” – a compromise we made for the initial launch due to technical constraints. 

Who should be our target audience to drive the most conversion? 

Repetitive value props caused fatigue

As if the value prop was not complicated enough already, we used the same copy throughout the entire flow. 

Is that really something we had to repeat every single step, or should we adjust what message to get across accordingly?

Overpacked content produced cognitive load

As soon as people tapped the Magic Link, we couldn’t help but overpack information from the get-go. It reduced the customers’ ability to comprehend what we tried to communicate. We did not give them good reasons to stay with us or continue.

Zoom in – a deep look at emotions

A user study before September 2019 launch included valuable observations and customer quotes of how they felt about the Magic Link provisioning flow. Our post-launch analytics validated those emotions and the data insights were related.

Zoom out – where people are physically

When I pulled the screen flow, customers’ quotes, and their emotions together on the same journey map, we got more clarity of the connection between what people see and what they do. What we showed customers needed to correlate with their location. 

Overall strategy

Reduce friction

Check in with the Engineering teams – “What does it take to remove those interim (extra) steps in the launched experience?” To make the “Magic Link” truly “magical”, we must reduce friction as much as possible. 

Customize the experience

Different customer segments have different rationales to get the new credit card. We need to customize our messages and encourage people to use the card based on the context. That means we speak to them differently depending on who they are and where they are during their shopping journeys.

Design highlights

Confirm sooner to avoid fatigue

It turned out that the total steps from card approval to using it couldn’t change. Yet our Engineering Lead assured us we could add the card right after customers tap on the Magic Link on a technical level! That meant the UI could show the “Card added” confirmation message sooner. It could make adding the card feel more a breeze. Consequently, using it would feel easier, too.

Give back control to customers

Instead of making it mandatory for customers to open Walmart Pay also confirm adding the card, we would give people control. They could choose to use the new card in Walmart Pay now or later. This option can reduce the stress for new customers. They could lack the mental bandwidth and patience to set up and learn how to use Walmart Pay at that moment.

Content matters – succinct and contextual

Collaborating with a content strategist is of big help! We managed to reduce the copy and improve people’s ability to digest information at each step. And we made the communication more on-point – when customers need to know, and what.

Give customers "how-to" when they need it

The moment to confirm “Card added” is not necessarily the best time to shop. Let alone learning how to use Walmart Pay. When it is time, we assist them with instructional copy alongside delightful illustrations.

Key takeaways

Zoom out before diving in

Data can show if something is working well or not. But it can never tell why. Pinpointing one screen or two is not going to cut it. Go beyond the surface and use artifacts to show a bird’s-eye view before diving deep into the weeds. It helps me to look at designs with a more strategic lens.

Obsess about words

One of the things I’ve noticed over the years, sadly, is that most of the time people don’t read. This appears to be more evident in mobile apps. Stunning visuals and amazing animations can make a digital product pop, but they cannot do that alone. Being able to collaborate with content strategists is a luxury. It inspires me to appreciate more the beauty and magic of words, which can bring products to life.

Get a set (or more) of fresh eyes

Critiquing my own design is scary. I needed to check my ego at the door and borrow teammates' fresh eyes for peer reviews. It's okay if I didn't get it right in the first place. A healthy dose of critiques can poke holes and beg tough questions that I may have ignored or have not thought hard enough about.