Podcast creators upload their episode audio files in the Episode Creation Flow. To fully understand the creator’s task flow, I created this journey map.

First, I wireframed ways creators could signify where they want their promotional audio inserted. These wireframes show 3 different directions.

I tested these 3 wireframe prototypes with audio engineers and new creators to make sure these interactions match existing patterns.

In this high-fidelity example, creators can easily place, edit or move their marker in the waveform. The waveform is locked to the bottom of the screen to match audio editing patterns.

Creators can add their promotional audio in the podcast show settings.
‍
To ensure users know about this new feature, I also created in-app promotional cards to onboard creators to this new feature.

I worked with our engineering team to implement tags to track user interaction on the waveform. We set up signals to track dead clicks and frustration signals.

After launch, our support team received complaints that announcements weren’t playing on episodes. After upload, creators would immediately go to a podcasting app to listen for the audio promo. When they couldn’t confirm the update right then, they were frustrated.

On the backend, audio insertion takes time. Hundreds of files need to be distributed and generated at different bitrates. I realized that there was a gap between user expectations and experience.

To clearly communicate audio insertion loading times, I added more context to the success alert when creators upload and submit a file, communicating that their audio would take up to 30 minutes to appear.

I also added a “distributing” badge to the audio file block, which disappears when all files are updated. Also, the 2 buttons from V1 are replaced by a quick menu with clearer actions.