Content is the most important part of Yik Yak. It's what users come to the app for. That audience is what creators stay for. To that end we wanted to give users tools that would allow them to more deeply express themselves through their content. We wanted to bring rich media to the app in the form of photos, videos, gifs, and links. It was one of the most consistent, unsolicited requests from users in almost all user interviews. "When are we getting gifs?" "Are you ever going to allow links in the feed?" These were questions we heard pretty consistently.

User sentiment was excellent, but the larger question was what impact should we expect on our key goals with the introduction of rich media as content? Our hypothesis was that this content, being of a higher quality, would solicit more interactions. Interactions (voting or replying to a post) were the lifeblood of a healthy feed. Success for us would look like a sustained lift in both voting and replying app-wide.

Content Quality

While we wanted to give users what they asked for, there was something we needed to consider: perception. Most image or gif content is primarily humor based. Humorous content is awesome, but around this time we were optimizing for utility-based content as we believed it would foster more meaningful one on one connections between users. Over time, through data, research, and user interviews, it became clear that our hypothesis on the value of utility-based content was off the mark. What users valued in the app first and foremost was humorous content. The kind of stuff they could sink a couple minutes into if that's all they had. Utility-based content was valuable, but it wasn't what users were coming to the app for primarily.

We began rolling out rich media in the form of GIFs in Chat. We contained the feature to this one dimension of the app for a few reasons. There were several product reasons but the only one I can share was observation. We knew this feature was highly requested but we wanted to know what actual usage looked like.

• Were users relying mostly on the trending gifs or were they searching for their own?
• What topics were they searching for?

One of the more interesting bits of data we wanted to collect was how often users would initiate a chat with a gif. Chatting was a high-friction action for users but it correlated with excellent retention. Because of this there was an ongoing mission to reduce that friction. The hypothesis was, "If we surface Gifs as a tool to begin new conversations, would we see a lift in chat initiations?" This feature came fairly late in the lifespan of the app so we were never able to validate the hypothesis.

All in all though this was a ton of fun to put together. Building out an input for variable content types is way harder than it looks. I can see why more products elect to go with the single content-other-than-text button and leverage native components like Action Sheets (iOS) or bottom sheets / Dialogues (Android) to surface what those other content types are.

In our case we wanted to explicitly communicate what those other content types were for discoverability. We wanted users to know they could post more than just text and we didn't want to obscure that behind a second tap in the short term.  Here's a look at the final product: