Osmo Words

Increasing engagement and retention with home users


Product, UX/UI, App

Osmo Words is a game that encourages kids to look at on-screen clues to guess the secret word, then use letter tiles to spell out the word. While Words was popular in schools, it had the lowest retention of all our games for home users - retention defined as the percentage of users who come back to play day after day. So we wanted to examine ways to increase engagement and retention for our home users.

For this project, I collaborated with game designers, learning researchers, and engineers. I owned everything design-related on the product - from wireframing and defining the user flow to final visual design and engineering handoff.

The behavior of users across our games during their first 40 days. Words is our lowest performing game in terms of retention - only about 40% of users come back for a second day at any point during those 40 days.

Challenges to retention

Whether a kid wants to play Osmo are a function of their answers to, “Can I do this?” and “Do I want to do this?” If both are yes, then they are likely to engage. However, we realized that there were challenges to both of those questions.

Do I want to do this?

Can I do this?

Playing Words alone isn’t as interesting for most kids - it can be less challenging and less exciting. Playing against friends is often more engaging, but kids don’t always have another person to play with. 

Albums start with the first letter missing and then move to guessing full words. However, a 5 year-old is capable of the easier levels but can't move on. There’s a layer of progression that is missing.

Developing a progression

We worked with a learning researcher to develop a better progression throughout the game. With the existing content, the game quickly became too difficult for 4-6 year-olds, so we aimed to develop a fun progression that was tailored to kids’ literacy development.

Increasing engagement with competition

Kids are most engaged with Words when playing with friends. We wanted to recreate that excitement, competition, and energy through playing against the computer. We first explored the most minimal ways a kid could play against a computer. 

Computer has no visual identity

Computer has an avatar

Neither of these options emulated the same excitement of playing against characters. We then decided to observe kids playing Words, and noted the instances where kids would get excited. We then built a prototype with these moments using stock animated characters


Initial prototype testing showed that playing with characters did increase the excitement and suspense of playing Words. Kids would say “We both got U at the same time!" or "What! His G came before mine one second!"

Developing Words Adventure

Through testing, we validated the concept that playing with characters is engaging. So we then developed the full experience. We developed both cooperative and competition modes, because through testing, we realized that younger kids didn't like competition as much. 

Initial wireframe of competition flow

Early draft of the competition flow

Exploring ways to show point totals

Full flow for Words adventure 

Impact and reception

The new Words competition mode successfully increased engagement via number of hours played within the first 30 days. Additionally, anecdotal feedback indicated a better overall experience, especially for younger kids. One parent noted - "A drastic improvement. Much more user-friendly for children just learning letters, phonograms, and beginning to decode words."