← All works

Intuit Mailchimp • Surfacing pricing plans for 11% increased conversions

Growth experiment

Internship

Shipped  

Overview

During my 6-month internship at Intuit Mailchimp, I worked on 2 hypotheses to increase customer acquisition. I contributed to design strategy, wireframing, prototyping, user testing, and visual explorations. In the end, 1 of the hypothesis was validated by A/B testing and shipped, significantly increasing paid conversions and activations. The project will introduce the process reaching a successful hypothesis.

My role
Product design intern

Project time
2 months, 2022

Collaboration
Design manager, Product, marketing, content, research, engineering, data, visual teams

Outcome
2 months, 2022

⏺ Problem

Significant decrease in ease of choosing the right pricing plan

As Mailchimp evolved from an email marketing platform to an all-in-one marketing solution, we have identified a notable trend: a statistically significant decrease in ease of choosing the right plan.

/ Goals

Help users
Understand what's offered in each plan and which offerings suit their needs

Help Mailchimp
Create effective acquisition funnels that increase activations, paid conversions and revenue

/ Metrics

Qualitative data
Pre-launching user testing

Quantitative data
30-day A/B testing

⏺ Research

Forming a hypothesis

1. Understand quantitative data

The pricing page itself has been improved to help users understand what's included in each plan. However, only 9% of the monthly 80k traffic goes to the pricing page which is the only place that explains what's offered in each pricing plan.

2. Competitive analysis

25% of the 90 competitor websites audited display pricing modules on pages other than the pricing page. I compiled the research into a presentation deck and shared with 30+ stakeholders in a meeting to advocate for the hypothesis.

⏺ Hypothesis

Increase visibility of the pricing plan by introducing the module on pages with the most visits

If we increase visibility of our paid plans on the homepage and key features pages, then we will increase activations and paid conversion because prospective customers will have a better understanding of our monetization model prior to account creation.

⏺ Initial design POV

Feature page. Tailor the module to those with interest in specific features

Many users explore specific features on dedicated pages. However, the current feature page doesn't show which plans include those features.

Home page. Simplify the module while maintain its intrigue

The module's information needs to be simplified for the home page where other messages also need to be conveyed.

⏺ Key feature page module

HMW tailor the module to those with interest in specific features

We could even track user activity (if feasible) and dynamically suggest pricing plans that include the features a user has viewed.

A. Tailored recommendations

On feature pages, visually emphasize the pricing plans that
offer the showcased feature.

B. Highlight the features

Highlight specific features introduced on the page to reduce cognitive load when comparing plan. The module is also made scrollable to reduce page length.

⏺ Plot twist

Consistency over customization under time constraint

Design: New and effective patterns

The team championed the design solutions due to their effectiveness and novel approach. However, we decided not to implement the customized modules for this quarter due to the following reasons:

Data Science: Hard to validate the outcome

Due to the relatively low number of visits on the key feature pages, it would take a considerable amount of time to obtain results from A/B testing.

Engineering: Tight timeline

With a dozen key feature pages, the design necessitates unique content for each, posing challenges for maintenance if future changes occur. And we faced a very tight timeline.

⏺ New goal - Universal module

How might we condense the information while maintaining its intrigue for a universal module?

The module's information needs to be simplified for universal use across all touch points while clearly explaining pricing structures to the audience. In addition, the module should also spark interest, encouraging visitors to explore further and ultimately sign up.

A. With feature lists

Allows customers quickly understand the key offerings to be interested
X Might be redundant if users simply use this module as a CTA to enter the pricing page

B. Without feature lists

Might trigger users to click into the pricing page due to lack of information and innate curiosity
X Might turn first-time visitors away due to lack of information

⏺ Testing Round 1

Option A wins(with feature lists)

Deciding what information to include in the home page pricing module was a challenge. To address this, we generated a list of questions based on the goals we wanted to achieve. Our researcher helped conduct 8 unmoderated tests. In the testing, we found that differentiation is more intriguing than lack of information even for earlier business stage participants who are not as familiar with marketing terminology/products.

Goal 1. Clarify

Measure. Comprehension of plan difference
Key question: How confident are you to choose the right plan?
Option A. 4/8 users are confident
Option B. 2/8 users are confident

Goal 2. Intrigue

Measure. Intent to visit high-converting pages or sign up
Key question: What would you do afterwards?
Option A. 6/8 would go visit the pricing page/create an account
Option B. 4/8 would go visit the pricing page/create an account

⏺ Iteration

Further clarify and differentiate the plans under constraint

Constraint: We needed to ship fast before the holiday season. The last iteration incorporated the engineering team's input in removing the interactive elements due to a tight timeframe. However, assessing whether it would suffice, testing with 8 participants revealed that it was confusing.
The pricing calculator

Meet constraints half way and avoid "dark pattern"

Before

Option 1

Option 2

Reduce confusion and better differentiate the plans

Before [With engineering constraints]

1. Lack of info makes it hard to differentiate the plans
2. Features names are difficult to understand without tooltips

After [Lifted some constraints]

1. Use check and cross marks to indicate plan differences
2. Add tooltips to explain the key features

⏺ Testing round 2

The improvements were validated, green lighting A/B testing

We conducted another round of testing with 8 participants, comparing the new designs to the old ones. The inclusion of cross marks and tooltips resulted in reduced confusion regarding the features offered and the distinctions between each plan. With the improvements validated, constraints were lifted, and A/B testing was given the green light.

⏺ Where I lent a hand

Visual revamp & responsive design

The site was undergoing a visual overhaul to target more established companies - "More mature looking and get rid of the Kale color". I explored the visual style of the pricing modules by collaborating with Mailchimp's in-house brand team. I also revamped 20+ pages across desktop, tablet and mobile devices.
Original 00 - Kale
Direction 01 - Pumpkin
Direction 02 - Cavendish
Direction 04 - Cornflower [Winner]

⏺ Outcome

Successfully passed A/B testing and deployed to all users

This marks the first instance of introducing pricing plans on both the home page and key feature pages. The 30-day A/B testing showed positive results. The new pricing module creates a successful marketing funnel to sign-ups.

↗ 10%

Activations

↗ 11%

Paid conversions

↗ 14%

Est. Revenue/user
(14M users)

↗ 10%

Pricing page visits

⏺ Takeaway

Gain a holistic view to advocate for best UX under constraints

Engaging in effective communication with cross-functional partners enabled me to grasp the diverse goals of various teams within a highly interdependent project. It was also crucial to determine when to compromise and when to advocate for user needs.
Presenting my work to everybody at Mailchimp in the end 💛
Let's create things together!
LinkedIn | Resume | yalaipang@gmail.com
Built in 2024 New York ©️ Claudia Pang