Tests hiding the large featured area at the top of Quest Home during a short campaign window.

The top Quest Home area controls which quest or reward gets the strongest first impression. The date gate suggests Discord is testing a temporary home layout for a campaign period.

Makes Group DM invites show who invited whom and a clearer join action inside chat.

Group DM invites can be confusing if it is not clear who sent them or where they lead. Clearer invite cards help users decide whether to join and understand when they need to ask for a fresh invite.

Clarifies that a Nitro gift is for one month in gifting copy.

Gift checkout should make it clear how long the recipient will receive Nitro. Clearer duration copy helps buyers understand the gift terms before paying.

Lets users fine-tune or combine the app, emoji, GIF, and sticker buttons beside the chat box.

The chat box is used constantly, so small control changes can be very noticeable. Users on narrow layouts or those who use fewer media buttons can reduce clutter, while frequent users can keep the controls they need.

Tests opening a global soundboard wheel when the soundboard keybind is held.

Frequent soundboard users can pick sounds quickly without first navigating back to the voice screen. This mainly improves speed for users who rely on keybinds.

Adjusts the video processing path for Windows PCs using Intel graphics during voice or video calls.

Video calls and screen sharing can behave differently across hardware combinations. This experiment is scoped to Intel Windows systems to validate a more suitable processing path.

Tests whether the desktop app uses a WebP asset for its startup splash screen.

The startup splash asset can affect the first moments of perceived launch speed and resource use. The code confirms that enabled is passed into the DESKTOP_TTI_SPLASH_USE_WEBP setting.

Extends the private profile flow that explains who can see profile details like bios and connected accounts.

Profile bios, connected accounts, and activity details are sensitive visibility surfaces. The code uses this experiment alongside existing private profile gates in UserProfilePrivacyNotice and AccountCoachmark paths.

Tests how sponsored content or ad CTAs are measured from load through external click and impression states.

maxStep ranges from -1 to 4, letting Discord roll out and validate deeper CTA funnel measurement in controlled steps.

Tests whether Steam timeline signals are included in game activity scoring.

Separating Steam timeline input lets Discord measure whether that activity signal improves scoring quality for game and clip experiences.

Tests a Windows voice-call setting that reduces echo when speakers are used instead of headphones.

Speaker calls can create echo when the other person’s audio is picked up by the mic. Splitting mix and auto lets Discord tune echo reduction with less impact on voice quality.