WORKED TOGETHER WITH EMILY YEH, JY LIEW

December 2016 - FEbrary 2017 @BLUESTACKS      

SAN JOSE, CA

Observations

  • On Twitch streams,  chats are always hard to read. There are always too many chats, which are also running too fast. 
  • It is even harder for stream to respond to viewers since he/she has to focus on the gameplay at the same time.
  • At a glance of the chats, the only thing that pops out turns out to be all kinds of emojis. 

Design Goals

  • Create a product that enhances stream engagement so viewers can feel more involved in the stream so they don't get frustrated when their chats get ignored. 
  • Experimenting with how we can attract users in Taiwan by introducing localized stickers/ emojis on Twitch stream, which is in line with the company's business goals

Initial User Research

DESIGN IDEA

As the first step of this product experiment, we decided to quickly test out a simple app that let stream viewers directly send stickers to streams that they are watching. We were hoping in this way, viewers can express their emotion in a fast, easy and vivid way, which is also easier for streamers to catch the message.

We decided on three types of stickers that we would use in our app

  1.  Standard ios emojis: people are already pretty familiar with these emojis. They can throw them on screen whenever you want. We decided to keep these small so it wouldn't affect the gameplay even when people are spamming. example:
 
 

     2.  Customized small stickers: we decided to create a bunch of stickers that are designed for Asian streams. They only show up on a certain area on screen. example:

 
 

    3. Full screen stickers: we wanted to test out another kind of sticker what would have bigger impact on screen. Considering that it might affect some of the gameplay, we were thinking that users should have limit use of full screen sticker. And in the future, they can even pay to unlock this kind of sticker. example:

 
 

WIreFrames

Brainstorming Session

DESIGN VARIATIONS

Design V1

Design V2

Design V3-1

Design V3-2

We started with really simple layout like categorizing by sticker character, by popularity, etc. Then we started adding more features to help users spam these stickers on screen to create a fun experience.

Design V4-1

Design V4-2

Design V5-1

Design V5-2

We discussed pros and cons of each design. And we agreed that as the very first product experiment, we should keep it MVP and would add extra features later on.

First ReleasE

There are basically two types of Pika users. One is streamers, who would configure their Pika as an overlay on top of the stream. The other one is viewers. By scanning a  QR code on the streamer's overlay, viewers can connect their Pika app with the live stream and interact with the streamer. Therefore, we designed a mobile app for viewers and a dashboard for streamers to configure their Pika overlay.

RESULTS & FEEDBACK

For the first test, we only released Pika to a small number of streamers in Taiwan for three weeks. We did see streamers and viewers enjoy using pika. But more importantly, we discovered many problems that we should continue working on.

Learnings.jpg

Second Iteration

Based on the feedback we got from the first release, we made the following main changes on the second iteration. 

 
!87, which means you are an idiot

!87, which means you are an idiot

!66, which means you are awesome

!66, which means you are awesome

 

RESULTS & FEEDBACk

It turned out that Pika V2 actually did pretty well. Without any marketing work, we surprisingly saw great traction and even more and more organic streamers. 

ONgoing Pika

Our team is actually still working on Pika, adding new features/events to enhance user engagements and stickiness. Here are some examples: 

more sticker options

!GG, good game

!GG, good game

!QQ, crying

!QQ, crying

!OO,  special sticker for lantern festival

!OO,  special sticker for lantern festival

 
 

We are still working on this project.

TO BE CONTINUED

BACK