Screen of Mailchimp Developer homepage

Slack

Enhance Slack's checkout experience by leveraging data-driven insights to optimize the purchase process.

Problem

The purchase experience is the beating heart of the business and changes in this area requires surgical precision. As such, this is a generally risk-averse space and success is measured by small signals from neutral to 1-2% positive.

When my team and I saw an increase of 2x more traffic to checkout, but conversion dropped from 40% to 23%, I knew there this area was ripe with opportunities for improvements.
Brainstorm artifact with stickies and screenshots of competitor's checkout flows with annotations
Cross-functional brainstorm board and competitive tear-down sample.

discovery procEss

I ran a cross-functional team brainstorm and surveyed other checkout experiences. Two key takeaways:
  • Too little friction can be bad
    While exploring a competitor's checkout process, I accidentally bought an enterprise subscription. Simplifying the checkout should prioritize user convenience without inadvertently promoting unintentional purchases, which could erode trust.
  • Identified gaps in our data
    We were missing data to inform our decision on which opportunities to pursue. Checkout data logging was outdated or unavailable. This made answering questions like how long does it take a user to enter and complete checkout and which fields sees the most drop off difficult.
I partnered with PM, data scientists, and engineers to fill the gaps in our data collection.
Guiding principles for the project

3 sections: reduce friction with checkout form pictured with 2 cogs, simplify checkout steps pictured with two lego blocks, and rescue abandoned carts pictured with a life preserver
Principles and strategic direction

approach

In anticipation of additional data, I established guiding principles to strategically navigate and guide our direction. Then, as a team, we identified opportunities we can tackle today and ranked its efforts from low to high.

Reduce friction with checkout form

Checkout was location-agnostic and required users to select their country and currency before the payment method element would load. A simple implementation of automatic user location detection helped an action from the flow.
  • Be a great host
    Implementing automatic user location detection not only streamlines data input but also minimizes the time spent on form completion.
  • Don't make me think
    With location pre-loaded, payment form is displayed from the start, reducing user cognitive load.
cropped image of checkout flow with a selection of countries
cropped image of a checkout flow form
image of a checkout flow form optimized
Before: User selects country, then payment form loads.
After: Form is pre-loaded from the start

Simplifying checkout steps

There are two distinct steps to checkout: the form and the confirmation. In the confirmation page, we observed a 7% drop-off. It lead me to question: 
  1. Is the presentation of the review step confusing?
  2. Is this too much time to second-guess the purchase?
  3. Is it necessary for returning users to go through 2 steps?
What if we eliminated the second step entirely?

Removing an entire page uncovered relational complexities between each form field. I worked closely with the engineers on each step to ensure each wrinkle is ironed out with precision.

As a result, the experience helps users understand when they've completed a step on the page, prevents them from moving forward when it's incomplete, and clearly indicate to the user when it's time to confirm purchase.

An added benefit to a one-page checkout is allowing returning users to purchase with one-click.
simplified checkout flowsimplified checkout flow
After filling out the checkout form, users are taken to a review page to confirm purchase.
simplified checkout flow
Streamlined checkout flow reduced 2-pages to one, helping keep the momentum on users' intent to commit.
simplified checkout flow
Returning users will have information pre-filled allowing them a one-click purchase experience.

Rescue abandoned carts

On the other end of the purchase journey, what can we do to help users who abandoned their carts come back to complete purchase?

Our data correlates users abandoning carts and signals intent on returning to complete purchase, but we didn't know what motivated this action. We needed qualitative data to better understand this behavior.

In a team brainstorm, we gravitated towards email as a channel of reengagement.

🤔 But Slack's value proposition is to replace email.

I pushed our team to think beyond email and put in the extra effort to implement an abandoned cart nudge and survey in-product.
brainstorm artifact with stickies and a screenshot of Slack's website with the headline, Is Slack better than email?
Brainstorm artifact and Slack's value proposition
Slack product image with an in-app notification
For high-intent users, we'll serve an in-app notification to nudge them back into the checkout flow.
Slack product image with a survey notification
For lower-intent users, we'll serve an in-app survey to better understand what is prohibiting them from committing.

Results

The experiment for one-page checkout resulted in +3.1% increase in conversions to a paid plan vs. control. Based on the results, estimated revenue projects an increment of $6M per year. All self-serve customers now experience the new one-page checkout flow.

In addition to a simplified checkout experience, this project also revamped data logging for improved usage monitoring and product health, and introduced an in-app survey to seamlessly integrate user insights with data analysis, deepening our understanding of user experiences.

role @ Slack

Lead Product Designer
My role encompassed collaborating with cross-functional teams, establishing design processes, coordinating with various teams including business operations, marketing, and feature teams. I actively engaged in sharing updates and soliciting feedback from leadership.

As a member of the design team, I also participated in mentorship and contributed to the design team culture squad.

---
Project completed in 2022 at Slack.

Role & Team

Lead Product Designer
Responsibilities included research, product design strategy, design details, and mentor to junior designer on the team.

Design team also consisted of a design manager and junior product designer. Product manager, 3 engineers, 3 content strategists, and an account lead made up the rest of the team.

---
Project completed in 2020 at Postlight.