Rethinking product onboarding for a new user base.
12 Week Independent Student Project (2021)
Role: Product Designer, UX Researcher
Published on UX Collective:
Using Video Games to Learn about Product Onboarding
To better introduce a complex, technical product like Discord to a new user base, I sought to reorganize and provide a more focused flow to Discord's product onboarding model in order to elevate the experience of using Discord itself.
To do so, I reframed Discord's onboarding into three specific stages that can be thought of as onboarding "playthroughs".
Onboarding Playthrough #1
The first playthrough is focused on getting the first-time user to the first milestone goal of simply talking to their friends in a voice channel. This initial goal affirms Discord's functionality and establishes a baseline for users seeking to use the platform at its most basic level.
UI elements such as spotlighting and tooltips are utilized to create a more linear guide through the first stage of onboarding.
Onboarding Playthrough #2
Once the first user objective is reached, onboarding can begin to expand user knowledge about Discord's core product, developing the user's understanding of what a server is.
The onboarding experience begins to start moving away from linear UI elements such as spotlighting, and adds new more unobtrusive elements such as clickable mission markers and banners.
Onboarding Playthrough #3
Given that server knowledge is established, onboarding now sits at a stage where learning and discovery should be more intrinsic, non-linear, and completely up to the user. In this third stage, the user has now ideally downloaded the native app and is now embarking on their own onboarding discovery, shaping the experience to their own needs.
In this playthrough, there's no need for extra UI elements and overlays, and components are essentially built to have integrated onboarding functions. Any extra elements should be indistinguishable from the core product itself.
It's in this stage that users start shaping Discord to their own personal preferences, and the Discord of an experienced user start to take form. At the perceived "end" of the third playthrough, the user's comfort with the product should be at an all time high.
Even among my own gamer friends, people would initially join servers that they were invited into, but would never return. In some cases, it was a lack of need or interest. But for many, it was because they didn't know how to start their own calls, or mute, or create their own channels, etc.
Because of how Discord's current product onboarding is set up, it wasn't effectively teaching new users how to actually use the product in the context of prebuilt servers. Most of Discord's most valuable onboarding came when people created their own servers, which people weren't doing for their first time in the platform
For the sake of onboarding, Discord's user base could be broken down into two specific types. One side consists of Discord's traditional gamer user base, while the other takes into account the new, inexperienced user base.
For "server rookies", the typical starting point for most of these users is a friend invite, with most coming over to Discord to join pre-established servers. We can reframe the onboarding journey to reflect this, with each user type representing the beginning and end of onboarding.
Knowing our specific user base and the journey that the server rookies go through, an initial onboarding flow can be sketched out. This can be shown with linear lo-fi sketches, broken down into three stages that each reach a respective onboarding goal.
Two different forms of user testing is used to optimize and make sure the onboarding functions as intended.
First, A/B tests with specific Discord onboarding components are conducted, which help distinguish which UI elements are more useful.
Second, I wanted to test each onboarding stage as a singular test, so each of the three onboarding playthroughs were tested individually to ensure that they could reach specific user milestones as stand-alone entities.
Once specific corrections, tweaks, and general tinkering is complete, final high-fidelity prototypes of Discord's revamped onboarding are created.
Designing for onboarding is a continuous process, and is something that I'd love to continue to explore more. Even within Discord, there's still more to consider, ex. how onboarding can evolve in relation to future updates and patch notes.
Beyond this first exploration, by understanding product onboarding itself, we can begin to innovate on the core product itself. For example, what if there was a new product better suited to the pandemic-inducted user base - maybe make a Discord mini for browsers?
(side note: if you're a Discord designer, I'd love to talk to you about this speculative project!)