To accomplish this, I:

• Adapted the flows to our content design system and our voice and tone framework

• Established and documented new content design rules to ensure consistency across the feature

• Considered the entire user journey of MyFitnessPal and Intent, from onboarding to subscription renewal

• Led reviews with Legal to validate product design for third-party integrations and ensure compliance early in development

• Partnered with Product and Lifecycle Marketing to collaborate on a suite of Premium+ upsell experiences leading up to launch

• Designed content for each of 7 distinct user journeys, from free trial experiences to Premium+ upgrades

• Spent all day, every day in Figma with our lead product designer and engineers to create the experience flow by flow

Meal Planner 1.0 launched as part of a new price tier (Premium+) in early April 2025 in the United States, Canada, UK, Ireland, New Zealand, and Australia. In October, Meal Planner received a special mention inTIME’s Best Inventions of 2025.

Build a custom LLM that could tap into users’ logging history to offer on-the-spot insights and suggestions as content cards on their home screen. Then train that LLM on our voice, tone, and messaging guidelines.

As the lead content designer on the Search team, I collaborated with product management, design, and engineering. To create the content portion of our prompt, I added our content and brand voice guidelines for our LLM to reference. Then I created a 10-scenario testing framework with high-quality content examples our LLM could follow.

After lots of testing and tweaking the prompt (and a few arguments with Claude), we found we could generate in-app content in our brand voice consistently at scale. For us, that meant less time editing, approving, updating content in the app, and more time to focus on content strategy.

Not only is this possible, but it worked faster than expected. With the right framework and examples, we were able to strike a balance between controlling the output and letting AI take the wheel. We were able to show the right message to the right user at the right time—something that had previously been impossible with traditional content design and segmentation.

We used AI very intentionally to solve this specific problem, not just to spam users with an overwhelming amount of content. Our work was just an experiment, but it unlocked some exciting possibilities that are currently in development.

Before making time-consuming changes to these tools again, we wanted to understand if the problem was user education and awareness. We suddenly realized that beyond onboarding, we did no notable in-product user education about the Premium tools. It was very likely this was a content problem, not a feature problem.

We started with a low-lift content test: a set of push notifications and in-app content cards about the Premium tools. With data science, we identified critical days and times in a new user’s trial journey where we might have the biggest impact.

As the lead content designer on the team, I created messaging with a “What’s in it for me?” angle. Users might already understand how our meal scan tool works. But do they know it’s a great option for logging discreetly when they’re out to eat? Or that voice log actually lets them log a whole day’s worth of meals at once? We focused on the second-order benefits users can enjoy. After all, it’s about them, not us.

We bypassed the usual process and ran this as a rapid no-harm test, and we got a quick signal: the campaign increased Premium tool engagement for trial users by 16% in just weeks, validating our hypothesis. Users may not have known why they should use the tools at all.

Feature adoption depends on continuous value reinforcement. Instead of jumping to update the tools, we guided user behavior with a simple content test and made a significant impact quickly.

Marketing’s first version of the campaign was rather transactional: get users excited about winning the prize, which was a discount on their Premium membership. While that’s very important, some of the story was getting lost. We were challenging users to log food daily not just to get the discount, but also to get in the habit of logging. Ideally, this will make them more engaged users, and more likely to convert in the long run.

As lead content designer on the project, I stepped in to offer a UX POV. I partnered closely with Lifecycle Marketing to give their three-week challenge a clearer progression. I helped reshape the narrative across channels with a strong through-line. I asked critical experience questions: What happens if users fail to complete the challenge? How do we build motivation and momentum week after week? How can we visualize and gamify progress instead of relying on standard reminders?

We evolved the campaign from a traditional marketing push into a product-led engagement experience. We fused lifecycle messaging with in-product behavior to help keep users engaged. Not only did we want them to be excited about the prize, we wanted them to be proud of their progress and get in the habit of logging, too.

When campaigns live inside the product, content design helps create continuity and momentum. Applying UX thinking to lifecycle messaging turns reminders into experiences that actively support ongoing behavior change.

We collaborated with our Lifecycle Marketing team to test our way into what worked best. We turned Meal Reminders into its own mini product team. This meant a rigorous A/B test schedule, building on successes, and cutting messages that didn’t perform well.

Starting from scratch, we saw a huge opportunity to target customer segments with relevant messages for their user journey and habits. We also mixed regular reminders with moments of surprise and delight to reinforce our brand voice. Selfishly, I had a lot of fun writing dozens of punchy messages in our voice, testing new angles, and taking our tired messages in a very different direction. We had a blank check to test the content we wanted as long as we kept the user tests to small groups.

Our first small test, which ran just one week’s worth of new messaging, showed a 3% increase in engagement. And that was just the beginning. As a content designer, there are few experiences as satisfying as getting to lead a content-first strategy from the ground up and hearing the words “Have fun with it.”

The company was changing rapidly, and we didn’t want it to lose its soul. Working with creative leadership, I organized regular cross-functional Figjam workshops with our product and marketing writers to create concrete voice and tone guidelines. As a council of word nerds, we put our heads together to debate the big questions: who are we, what is this place, and where are we going?

We used the output of our workshops to create a series of high-level personas that tell the story of who we are on our best (and worst) days, whether in emails or in the app. MyFitnessPal is a beloved legacy product, and it’s always had a voice, or at least echoes of a voice, mostly shared through organizational folk knowledge. By giving this voice a name and adding aspirational guidelines in an organized framework, we started to bring the brand to life and define what makes us unmistakably us.

It’s easy for marketing and product to stay in their own lanes long enough that voice and tone can evolve differently in parallel. Proactive cross-functional collaboration and documentation can help us all speak the same language and give the brand a stronger point of view.

It’s not that nobody wanted to do this documentation work. Everybody was understandably too busy. But when we prioritized this work and made the time, great things happened that benefitted us all in the long run.