Flo Health on how its content drove daily app usage with pregnant mothers
Flo Health has won the Data and Insights and the Healthcare and Pharma categories at The Drum Awards for Content. Here is the award-winning case study.
Flo Health is the most popular women’s health app globally. Over 300 million people have downloaded Flo, and 55 million people use it on a monthly basis. With over 120 medical experts informing Flo’s content, Flo supports women during their entire reproductive lives and provides curated cycle and ovulation tracking, personalized health insights, expert tips, and an anonymized community for users to share their questions and concerns.
Objectives:
In our ‘Pregnancy’ mode, we empower pregnant women with the information they need, when they need it. Our challenge was to revolutionize the way we provide this info, offering digestible and super-engaging Instagram-style ‘stories’ (rather than articles, as had been our focus), tailored to their specific pregnancy.
In order to create a more profound connection with our users, we needed to create an experience that felt ‘smart’, with all of the personalized elements auto-generated to appear without any extra input needed from the user. Users input their due date when they join Pregnancy mode, so we knew we could serve content based on their week of pregnancy. However, this meant that each content topic had to be rich enough to provide 40+ pieces of information. We also had the challenge of delivering this project (which turned out to be 334 stories, including anatomical diagrams, graphs, interactive elements, etc.) within just two quarters.
We measured our success in this challenge in two main ways:
uplift in free users taking out trials of our premium subscription (which have a high conversion rate) and, subsequently, paid subscriptions
uplift in the number of users opening more than one story for our existing premium users
Strategy:
Firstly, our Content team sat down with our Medical Advisors and UX Researchers to find out what our pregnant users wanted to see throughout pregnancy, and what scientifically-backed information we could accurately provide. We ended up dividing the planned content into three types:
Weekly baby growth updates
Weekly body changes information
Daily stories with a countdown to the due date and extra information, which we divided into seven types (one for each day of the week): ‘Question of the day’, ‘Swap of the day’, ‘Myth of the week’, ‘Secret Chat of the day’, ‘Win of the week’, ‘Reminder of the week’, and ‘Mood check’.
We decided to develop the first batch of stories for users in their first trimester (the first 14 weeks of pregnancy) and release this for testing at the end of August 2022. We ended up creating 161 stories — an incredible feat — with 14 weekly baby stories, 14 weekly body stories, and the rest made up of the daily stories.
After seeing the success of the stories, we wanted to make sure there were no delays in delivering the rest of the content to users. So for the second and third trimesters, we came up with an efficiency hack. We analyzed the first trimester stories’ performance then chose the top three daily formats to carry on with. Instead of there being a new daily story every day, there were three new daily stories each week — tied to when users started a new daily session, not a new day. With this plan, we were able to release trimesters two and three over the end of 2022 and start of 2023.
Evidence of research, analysis and application:
We based our work on many different streams of data and insights, both internal and external:
RESEARCH: Our 'Swap of the day' stories were taken from our Secret Chats community (an active in-app forum).
ANALYSIS AND APPLICATION: We identified the top things our pregnant users missed and gave them alternatives. For example, pregnancy-safe sushi and guidelines for safe fish consumption.
RESEARCH: Our 'Question of the day' stories were inspired by top SEO search results during each week of pregnancy.
ANALYSIS AND APPLICATION: This led to useful content like, ‘Is having diarrhea at 7 weeks normal?’
RESEARCH: Content editors collaborated with designers to come up with a selection of attractive yet simple proof of concepts. We then carried out UX research to get real user feedback on what they liked or didn’t like.
ANALYSIS AND APPLICATION: Overall, users found the stories useful, informative, trustworthy and perceived them to be backed by medical experts. Based on answers around credibility, we decided to highlight our various sources of data e.g. personalized data, general population data, and medical research.
ANALYSIS AND APPLICATION: Unless there was an interactive element in the story, users didn’t necessarily realize information was based on their personal data. Therefore, for each of the weekly body change stories, we started with an interactive yes/no question (for example, on the topic of nipple changes, ‘Have you noticed this?’) and the answer was then tailored to the user’s response.
ANALYSIS AND APPLICATION: When directed to further content at the end of a story, users were more intrigued to click when they were told the content format, so we added this.
Results:
This turned out to be a huge project. In total, we delivered 334 stories: 161 autogen stories for trimester one, and another 173 for trimesters two and three.
Initial proof of concept results:
Rolled out to 30,000 users
For free users, we saw +80% in trials
For premium users, +30% in the number of users who opened at least 1 story
We then carried out UX research to test user reactions. Once we’d made adjustments in line with feedback, we rolled the first trimester stories out to a further 200,000 users. We then did the same with the other trimesters.
Total results (all trimesters):
+20% trials per day
+20% trials per week
By trimester:
Trimester one stories increased premium subscription trials by 1,600+ a month.
Their average click-through rate was 11.88%, which is up to 1.7 times higher than existing content.
Trimester two’s weekly stories increased trials and subscriptions by 44%.
Trimester two’s daily stories increased the number of users who went to start a subscription purchase by 11%.
Trimester three’s weekly stories increased the number of users who went to start a purchase by 20%.
User feedback:
When we first launched the experiment with the new stories, we immediately had users writing into our Support team saying their friends had these really cool stories and they didn’t, wanting to know how they could see the new content. Users also posted positive feedback on our social media channels, with some even posting screenshots from their Flo app against their real-world body changes, such as the appearance of linea nigra (the brown line that can appear vertically on pregnant women’s abdomens) – see attached image.
Ongoing results:
When the stories were rolled out to all 480,000+ monthly active Pregnancy users, the results stayed strong. The weekly baby growth stories have done particularly well, and are some of our highest performing content in the app. They have a whopping 61.24% watch rate, with 0.75% of users starting trials after watching.
Annie O’Leary, Chief Editor at Flo, said: “At Flo, we put our users at the heart of everything we do, which is why projects like this where we act off the data and insights we get from those users, is so successful. This project is a great case study that we use internally to help our creative functions see metrics as the fuel for their imaginations and the innovations they deliver. I hope it can help serve the same purpose for others outside Flo, too.”
Explore frequently asked questions
Advertisement
Advertisement