Flo is the #1 women's health app worldwide. Over 300 million people around the globe use Flo as their ovulation and period tracker app, fertility calendar, and pregnancy assistant.

The project required close collaboration between various departments, including Product, Medical, Legal, Marketing, and Engineering. The Product team comprised a Product Designer (myself), Content Designer, Product Manager, and UX researcher.

Sexual and reproductive health conditions are common, but hard to diagnose. Endometriosis, uterine fibroids, and polycystic ovary syndrome (PCOS) impact a significant number of women and individuals who menstruate globally, with estimates ranging from 5% to 40% of reproductive-age women. Currently, it can take up to 12 years to diagnose these specific conditions, subsequently contributing to health complications and increased healthcare costs.

At Flo, we wanted to give women something they often lack: awareness. Symptom Checker highlights possible signs and patterns, empowering women to notice what matters and to make informed decisions about their health.

During the discovery phase, we explored the topic and created a simplified version of the feature as a proof of concept. This early version became our starting point for further design and research. The design process involved numerous iterations and user testing. Once we had a clear vision, we established priorities for iterative development and tested each increment internally. Finally, when we launched the first public version, we measured metrics and listened to user feedback (spoiler: the result was great!).

Such a project is not an easy thing to do, taking into account potential risks when it comes to people's health. So while doing discovery research and diving deeper into the topic, we created a very simplified (and quite straightforward) first version of the Symptom Checker.

To make sure the Symptom Checker performed as intended, we validated the algorithm through an independent clinical vignette study, where general practitioners created and reviewed dozens of simulated patient cases before running them through the tool. The study was later peer-reviewed and published in JMIR mHealth and uHealth (2023).

The lesson was clear: by connecting the dots — combining symptoms, context, and clinical evidence — we could create a tool that highlights what matters, while keeping the responsibility of diagnosis where it belongs: with medical professionals.

Turning medical complexity into something users can easily understand and trust is never simple. For any product team working in health tech, this process raises a set of recurring challenges:

• Balancing clarity and accuracy — how to simplify medical information without losing its scientific grounding.
• Maintaining empathy — addressing sensitive health topics with care, ensuring that information supports and reassures users rather than alarming them.
• Defining the right data scope — deciding which inputs to collect without overwhelming users or overcomplicating the experience.
• Protecting user privacy — ensuring that sensitive health data remains secure and anonymised while still enabling meaningful insights for symptom analysis.

Based on the initial version, its user testing results, and findings from the discovery, we started mapping the user journey to find out opportunities for the best experience.

This step included an enormous number of ideas and design iterations for every single piece of experience: first touch and entry points, assessment and its results, symptoms list and statuses, a user needs specific scenarios, etc. And to make sure we were going the right way we were running user testing to validate our hypotheses and ideas.

The Symptom Checker's primary access point is the widget located on the Today screen. It acts as a crucial indicator of the current status, alerting users to potential issues by analyzing and interpreting the data they've submitted related to cycles and symptoms.

Some symptoms are critical while some are not. If the tone is too alarming, it can lead to stress or even avoidance. To strike a balance, we used visual hierarchy, highlighting the most important points, a supportive tone with language that is informative, calm, and encouraging rather than frightening, and provided context to ensure users understand that one sign doesn’t indicate a condition, but combinations may be worth noting.

The initial version of the Symptom Checker emphasized clarity and transparency, presenting a direct list of assessments and diagnoses. This decisive design showcased the full features up front, building trust and empowering users from the start.

We built a chatbot that guides users through a simple conversational flow, collecting key details about their experiences and sending that information securely to our backend service for analysis.

Since recalling symptoms over the past year can be tough, we analyze users’ logged app data to detect unusual cycles or persistent symptoms. Combining these insights with self-reported information improves assessments and lessens reliance on memory.

Where possible, we added a signs and symptoms visualization so users can easily review their history, check for abnormalities, and verify accurate symptom detection.

Once the information is gathered and analysed, the next challenge is deciding how to present it back to users. The goal was to highlight meaningful combinations of symptoms without creating unnecessary fear or implying a diagnosis. 

Together with this assessment’s result, the tool generates a symptoms checklist. It shows which of the user’s own signs overlap with those described in medical guidelines, making the reasoning behind the result transparent. The purpose of this design is simple: to give women something they can take to their doctor. The checklist is meant to support, not replace, the next medical conversation.

Summarizing, here users can read about conditions in detail (about, causes and treatment), see the symptom checklist, get education from the Flo library, and connect with others in the Flo community. All of these support users in their awareness, education, and social journeys.

Finally, a promo story was designed to support the launch of the Symptom Checker. We tested several options of storytelling before finding the best way to catch users' attention and explain the value.

Since its launch, the Symptom Checker has reached millions of people around the world (Flo’s Symptom Checker is not available in the United Kingdom, European Union, or European Economic Area.). More than 1.8 million users have already completed assessments and learned about their results. For many, it was the first time they realized that their symptoms — the fatigue, the irregular cycles, the pain they had normalized for years — might be connected to an actual health condition.

Among them are at least 600,000 users living in countries with limited access to reliable health information, where the Symptom Checker is available free of charge as part of Flo’s Pass It On campaign. In regions where access to health care can be limited, this kind of information can be life-changing. Awareness becomes the first step toward care.

In our follow-up survey, 28,000 users reported that they spoke with their doctor after completing an assessment, 17,000 said they got tested for potential conditions, and 4,500 confirmed a diagnosis.

We’ve seen how knowledge itself can be empowering. For some, the assessment brought relief, a sense of understanding what had long felt uncertain. For others, it was the motivation they needed to finally talk to a doctor. Instead of searching the internet for answers or doubting their own experiences, they could walk into an appointment with clear, structured information about their symptoms.

The Symptom Checker isn’t just an algorithm. It’s also a carefully designed experience that respects both the sensitivity of women’s health and the responsibility of giving health-related information. Designing for healthcare isn’t like designing for any other product. Every detail carries weight.

• Words matter — the difference between “symptom” and “sign” can change how someone reacts.
• Tone matters — supportive, neutral communication builds trust, while alarmist language erodes it.
• Responsibility matters — good design in health is not just about usability, but about ethics.