patients like me

PRODUCT OPTIMIZATION:

A social media outlet for people with chronic illness

Role: Product Design Lead & Strategist
Company: Patients Like Me

01 Context

PatientsLikeMe (PLM) is a social media platform for individuals with chronic illnesses. It serves as a community forum for people’s shared experiences, provides curated information about conditions, treatments, and provides tracking tools for the day to management of one’s own wellness journey.

02 Problem

In a survey designed to capture awareness, more than 10% of respondents said they’d never heard of PLM. They didn’t know what we did, our value proposition. The logged-out pages had a high bounce rate; the sign-up flow was abandoned regularly, and had a high error rate.

Our directive was to increase PLM’s registered user base from 600,000 to 1 Million in 2019

03 Approach

Social communities thrive on participation with a trusted platform, so it was imperative to attract and retain a larger user base. Our goal was to increase the number of people engaging with the site/community to positively impact the Patients Like Me experience.

My team and I needed to absorb the institutional knowledge of our working partners. So, I facilitated design thinking workshops, and conducted informal interviews with key stakeholders including designers, researchers, marketers, and product managers to understand the various perspectives at PLM. Access to patients was restricted to specific times, so I also scoured the user interviews done by our in-house researchers.

My team and I chose to focus on three topics:

1. Increasing Brand Awareness

2. Improving the Stickiness of the Pages

3. Optimizing the Sign-ups Flow

Increasing Brand Awareness

While our engineering team was hard at work optimizing our SEO, we designed multiple A/B tests to discover what really resonated with our users. Was it a message about community? A personal connection to someone who’s been in your shoes? Or data to help them make informed decisions?

  • Option A:

    Community Support

  • Option B

    First-hand knowledge

  • Option C

    Data tracking

We found that connection with the community was the strongest pull for would-be members, but connecting with experienced individuals was a close second.

Making It Sticky

The logged-out pages of PLM were cluttered and dense. Like many products, content and features had been piled on layer by layer without much refinement or editing. We needed to simplify the landing pages and provide the most poignant information for someone new to PLM. We had to find the right balance between data, community, and messaging.

Optimizing Each Piece of the Site: Member Comments

Option A: Community Support

Option B: Connection with an Experienced Individual

Option C: Data for a More Informed Decision

Optimizing the Sign-up Flow 

70% of users accessed the sign-up page and its incarnations via mobile devices, so we needed to optimize the design to better support this usage. Additionally, we needed to examine our micro-interactions to reduce the error rate.

However, before conducting any testing, we needed to perform some initial clean-up to establish a solid baseline. To optimize the sign-up page for mobile devices, we narrowed the fields, switched from a two-column layout to a single column, and made the form fields more compact.

Example A/B Tests

Throughout each iteration, we developed a hypothesis and a method to test it. This involved not just an A and B option, but many options A, B, C, D, and E. We also took great care to establish baselines. Although the slides below show changes to multiple variables, we typically changed only one variable at a time. As a team, we negotiated how to manage both major design changes and smaller, more specific ones.

  • Option A:

    Email + Password + Username

  • Option B:

    Email + Password

  • Option C:

    Email + Password + Condition + Treatment

04 Impact

Year over year, with all of these changes, we achieved a 40% increase in signups. The breakdown is as such:

SEO + Brand Awareness Improvements

• Organic traffic improved by +50%

• Traffic to blog improved by ~104% post SEO and design changes

Value Prop

• Making design & content changes Treatment page resulted in 10% reduction in bounce rate

• 90% increase in viewing the signup flow and

• 10% improvement in signups

• The new user journey rated positively in the usability review

Sign-ups

• Conversion (viewed sign-up page —>sign-ups completed) + 5%

• Errors on submit (mweb) -15%

• Errors on submit (web) -8%