Live Polling
Why did we build this feature?
My team was responsible for developing features to increase event app usage and engagement amongst event attendees. We hypothesized that a Live Polling feature would help drive awareness of the app itself: event organizers and even presenters would promote the app and explicitly tell attendees to download the app and vote. By making polling a fun and rewarding experience, we might also create more overall engagement with other app features. We prioritized this feature because it met our criteria of:
- Having "built in" viral distribution, and the potential to drive app adoption
- Potential to drive increased app engagement and data capture
- It served to differentiate our product in the market: none of our competitors offered native, mobile polling capabilities
What we shipped:
There are many great products that focus exclusively on live polling and audience engagement. Our goal was not to compete with these standalone services, which are very feature-rich and highly customizable.
We distilled polling to it's simplest form and imposed a number of constraints in our design: polls are single question, intentionally short form, that allow only a single response. Since we were designing for mobile, we chose relatively short character limits and also limited the number of potential poll responses, so that users would never have to scroll to view all possible responses on their mobile screen.
To make Polls fun and rewarding, we invested time to in transitions and animations the make the feature feel fun and "slick" to use. We also tied Polls into our app's gamification system, so that event attendees could earn points by participating.
We designed Polls to appear in the app in two different places: global polls appear in the main activity feed, making them available to all attendees, and session-based polls appear in the session detail view.
What we learned:
We tracked a number of metrics after this feature was released:
- Feature Adoption:
- Did customers choose to use it? This gives us a baseline for understanding whether our customers perceive value in the feature.
- About 40% of all events created currently use Live Polling.
- Did attendees use it? This gives us an understanding of whether the feature itself is easily discoverable or whether there may be distribution issues.
- 23% of active users responded to a global poll
- 14% of active users responded to a session poll
- Did customers choose to use it? This gives us a baseline for understanding whether our customers perceive value in the feature.
- Impact on engagement:
- How does overall engagement compare between events using Polling and not using Polling?
- Attendees at events using our polling feature typically visited the app twice as often as those that didn’t. What’s more, on average, attendees at events that used polling also posted twice as many status updates as attendees that weren’t exposed to polls.
- Next, we wanted to better understand the difference between users that actually interacted with polls and those that did not for events with at least one poll. Dividing users this way highlighted a stark difference in engagement: we saw that attendees who participated in a poll in the app had 240% more visits to the app than those who didn’t. We also saw a significant difference in social engagement – attendees who took at least one poll posted nearly 5 times as many status updates in the app.
- How does overall engagement compare between events using Polling and not using Polling?