Phil Wall

When I joined Castlight we had three customers and poor registration. New users had to complete was a 28-click sign-up flow and input their social security number just to try the product.

Solutions:
#1 Get stakeholders on board: I presented our legal team with research and analytics data showing SSN was preventing registration but was told security is “non-negotiable.” Instead of arguing, I agreed and looked for solutions

#2 Design for trust and simplicity: we simplified the flow into 1 page with 60% fewer clicks, employer branding, coworker testimonials, and a Norton Verified Security badging. This reduced mid-flow abandonment by 38%.

#3 Remove the bottleneck via an SSO option: I knew many employers have benefit portals where users’ identities have already been verified. I recruited pilot customers and we built a secure single sign-on (SSO) registration. Identities are transferred and registered with Castlight in one click. No more abandonment problem.

#4 Embed Castlight in open enrollment (use existing channels): Each year employees are required to select a health plan. I asked Walmart to put “Register for Castlight” as the last step in their open enrollment this one pilot alone nearly doubled our user base.

"A must-have app." — TechRadar

"Sesame will change how you use your phone." — Android Unfiltered

Sesame connects APIs and native data sources to power a universal search for Android that is filled with deep links into your apps (Contacts, WhatsApp, Spotify, Slack, Reddit, Instagram, Netflix, and dozens more).

2MM+ users, >25% long term retention, and a top app in the Play Store. But we failed for 18 months before finding success.

Here's the story...
User testers loved it. I was expecting a big hit. When we released, no one cared and our retention rate was 3%.

Solution: "Crossing the Chasm" by Geoffrey Moore
• Found a niche of early adopters at Reddit.com/r/androidapps.
• Posted our app and promised if people gave feedback we'd push an improved version each week.
• At first, we got zero upvotes. After a few weeks, the community saw we were genuine and engaged. Our releases became the top posts in the community each week!

By hyper-engaging power users, we learned how to build a product people wanted. Eventually, a major influencer reached out to us for a partnership and we broke through to mainstream success.

That customer development process was the hardest and most rewarding accomplishment of my career.

I was the product lead for a B2B2C healthcare shopping tool that shows prices for all providers in your network. Too many users were logging in once and never coming back.

By studying the retention strategies of other platforms like LinkedIn, we were able to deliver our value prop to users without them needing to log in.

When a user went to a doctor, we would send them a "Care Team" email that showed their bill and what they could have saved at a provider with equal quality ratings. This improved our open rate 8X and led to a 2X increase in using Castlight to pick their next doctor (our #1 conversion goal).

My Role:
• Developed the strategy by conducting open-ended user interviews, then delivered solutions according to the "Hook canvas" strategy.
• Designed the UX and UI of the end to end solution and user-tested it.
• Conceived the key data features and helped design the recommendation algorithm.
• Designed and managed the building of the email targeting system, wrote the marketing copy, and A/B tested it

100,000+ Care Teams were created in our first year. Tens of thousands of users switched to higher-value providers, resulting in user and employer savings in the millions of dollars.

After reading a McKinsey white paper saying that 31% of new hires are regretted after six months and each mis-hire costs a minimum of one year of salary at that position, I researched how to improve hiring quality. I found there are decades of validated research that shows if you analyze which traits your top performers share, you can test for them with well-thought-out interview questions and assessments. The problem is very few companies have the internal experts to create this content,

My idea was to create a massive wiki of validated interview questions and correlate which answers predict success for different roles like senior software engineer or sales development rep.

If we could get companies to use this library, we could use statistical analysis and machine learning to increase our predictions over time. The math is all textbook organizational psychology.

ClozeLoop was a struggling 5-person startup that offered a virtual playbook for sales teams.

The founders were two well-regarded sales coaches and were confident there was pain in the space. But they weren’t getting traction with the product. I was brought in to fix that.

User testing showed that employees didn’t want another “source of truth.” They already had too much information in Google Docs. When they needed help, they would ask coworkers in Slack. Managers hated this, it distracted the whole company.

Instead of fighting inertia and trying to rewire behaviors, we took an HCD approach and oriented the product around the existing behaviors and solutions.

My Role:
• Started as a contractor and was offered a co-founder position after two weeks.
• Conceived of the solutions.
• Oversaw UX, UI, and visuals.
• Ran user tests, a design sprint, prototype build, and weekly iterations.
• Managed the project, owned the roadmap.
• Supervised an 8-week customer pilot.

Results:
• Found the problem: static content wasn't in the workflow.
• Pivoted to: an internal knowledge base that scrapes Google Docs, delivers answers via a Slack bot, and curates with Quora-style curation features.
• Paid customer in three months

Scenario:
Castlight started growing insanely fast. Our implementations were taking 3+ months and our support teams were overwhelmed. There was real anxiety that we did all this work to get off the ground only to stumble on execution.

As the lead product manager for the core app, I shifted my focus from acquisition to supporting our launch teams.

Solution:
• My first move was to tell our implementation and support teams "You are my #1 customer, tell me what you need."
• Discovery: Flew onsite to customer launches and worked as an assistant to our implementation leads as they rolled out our tools and dealt with customer complaints.
• Led the process as we wrote out each step in an implementation workflow, identified where the bottlenecks were, and what customers cared about.
• Organized a Kanban roadmap, designed the solutions, and dedicated my team's sprints to the task.

Results:
• We automated many implementation tasks.
• Built an internal set of tools called the "Configurator" that allowed the implementation managers to set up and launch customers without engineering help.
• Reduced implementation time and costs by >30% in six months.

Back in the days when Uber and Lyft would layer surge pricing onto your ride, it could be a hassle to pick the best option.

My developer partner Steve and I came up with a simple idea to hit their APIs and compare prices. We called it "Ride Fair."

I designed a stripped-down one-screen interface with some cool "delighter" features like a stylized Google Maps background. You could see the ETA and surge pricing rates for all nearby Uber and Lyft cars. One tap would link you directly to ordering the best ride. We built it in a couple of weeks and it was a hit!

Unfortunately, Uber and Lyft weren't too keen on the idea and had us shut down.