
A chat and voice platform for gaming friends and communities.
Discord grew by fixing friction in its core use case. Friends-in-games work lifted users from about 10 to 12 or 13 active days a month, driving some of its fastest growth since the pandemic. Non-gamers churned over jargon, so usernames replaced discriminators and friend-request acceptance rose. Lesson: fix the core loop, simplify for newcomers, and keep paywalls off the invite loop.
From Francesco Polizzi on Discord's redesign
When funnel tweaks stopped working, Discord redesigned the product itself. Non-gamers kept saying they downloaded Discord after a friend's invite, found it too complicated and deleted it. The growth team concluded standard funnel work could not fix that and chose to redesign mobile, where most non-gamers joined.
Pandemic waves (Among Us, a WhatsApp outage, crypto communities) had brought crowds of newcomers in 2021. The project began as a hallway talk between Francesco Polizzi and early engineer Stan about making Discord less bloated. "We've optimized everything that's possible here. We need to change basically where we're optimizing from."
Start with two weeks inside every rival app, not a strategy doc. Francesco cleared his calendar and used Skype, Telegram, WhatsApp, Signal, Messenger and iMessage on throwaway accounts, testing every feature and dumping every gap into one master list. Grouping problems into teams came afterward.
Every problem fell into quality, usability or familiarity. The team held that quality and speed were not at odds and aimed to reach the first thing they could see and feel together, which told them what to do next.
Replacing four-digit discriminators let non-gamers actually find their friends. Usernames like name#1234 meant newcomers typed a friend's name without the number and failed to find them. After the switch to unique usernames, the share of outbound friend requests accepted rose by what Francesco called a pretty crazy amount.
The team also split DMs out of the messaging tab to feel familiar. It kept the gamer term "server" but made servers feel less foreign. No exact figure was given for the friend-request lift.
A one-question survey justified voice messages despite a rival's failure. A PM whose nephew loved voice messages pushed back on an anecdote that the feature had failed elsewhere. A quick survey of a subset of users found 70 to 80% sent voice messages weekly, and the shipped feature lifted overall communication by a single-digit percent.
Usage was higher in South America and the Middle East, which Francesco reads as a generational gap. Surveys work when they ask about one concrete existing behavior: "How often do you do this thing that already exists somewhere else?"
Rebuilding buggy video chat raised video minutes between friends. Mobile video froze when users zoomed, so the team rebuilt Hangouts, a months-long and technically hard project. It increased video minutes between friends.
Francesco frames it as hunting the nooks and crannies where bugs pile up in a fast-scaled product, with video as one of the most intimate ways to connect.
A clearer profile tab leaned into Gen Z's taste for expression. The redesign made the profile tab clearly about how others see you and made customization easier to reach. That foundation later supported paid animated profile headers, effects and the shop.
Map where new and power users clash before shipping big changes. Discord's rule was that the new audience should be a superset of existing users and nothing should alienate them, yet the team never pinned down where the groups would conflict. Forcing everyone to change usernames caused a large uproar that still draws online noise.
Francesco says the disconnect with core power users was discounted more than it should have been, and no count of lost users was given. Phil Carter's advice for such moments: wait a few weeks past the vocal first-week backlash, then compare it with the lasting effect on metrics.
Discord grew fastest by removing friction for users it already had. In 2024 the team focused on being the best place for friends to play multiplayer games together, even though most such players already had Discord. Fixing bugs from Reddit threads and improving performance moved users from about 10 active days a month to 11, 12 or 13, driving some of the fastest growth since the pandemic.
Many users sat on the edge, using in-game voice or FaceTime instead. Midjourney's boom on Discord did not show up in the usage graph at all. Peter Sellis applied the same lesson at Snap, where Android performance work for existing users drove growth for years.
Define core value as a phrase, then turn each word into metrics. Discord's core value was talking and hanging out with friends before, during and after games, broken into metrics tied causally to daily return. It was used to judge roadmap items and to onboard new hires and data scientists, and the definitions went into Discord's growth playbooks.
Snap's version was the fastest way to share a moment with your best friends, which Sellis presented to every new Snap hire for seven years.
Find new bets by watching users bend the product to unplanned uses. Sellis cites Discord communities as users going through hoops to use the product in ways it was not designed for, one of two ways to spot expansion. The bolder way is backcasting from the founding idea with no tie to today's tech.
From Francesco Polizzi on Discord monetization
Charge for power-user perks, never for what drives friends inviting friends. Discord's growth loop is friends inviting friends, so Nitro charges for perks like larger file uploads and custom emojis that never block a new user joining a server. Display ads fit poorly because most chat happens among two to four people with little inventory.
At Photomath, a rival capped scans and damaged its core loop, while Photomath monetized something else and reached about a 100x outcome by Francesco's estimate. Paywalled features must still be genuinely valuable to someone.
Reward ads called Quests fit gamers better than display ads. Brands show relevant game content and users earn orbs to spend on profile customization. Quests became Discord's main monetization push after Francesco left.
The shift followed the realization that the COVID audience expansion was less lasting than assumed and gamers remained the core of the business. Francesco offers the ads view as an outsider.
Audit screens leading into payment, since one bug can cost millions. Tapping the Nitro button on the profile sometimes failed to load the landing page or took about 30 seconds. Fixing it was described as worth millions of dollars.
The figure is anecdotal, with no exact number or control group.
Subscriptions on messaging apps build slowly as cohorts form habits. Discord's subscription penetration followed the same slow, roughly linear cohort pattern as Snapchat Plus, because users must get used to daily use before they see the value. Sellis wonders if launching Snapchat Plus four years earlier would have eased pressure on ads.
"Time is your enemy with subscription products."
A shared mission plus named owners replaces most collaboration. As head of product, Sellis distilled the founders' vision into short repeatable phrases and gave each problem a directly responsible owner. PMs were pushed to make the calls the founders would have made a year or two earlier.
He argues collaboration makes a team as slow as its slowest node, and compares the model to a terrorist organization built on an ideology rather than a leader. Discord's chief people officer was "horrified and intrigued" by the framing.
Discord's founders took back UX from their least experienced people. They saw they had delegated the key product decisions to junior staff, so they flipped the pyramid and became the first and last mile on UX.
Mark Pincus uses this as an example of founders staying close to pixel-level detail for as long as they are the best player.
Spiky talent like Nikita Bier is directed, not managed. Bier reported to Sellis for a little over a year after his startup was acquired, bringing strong engineers. Sellis says you don't manage him, you only seek to direct him at problems where he is expert.
Sellis once pinged Bier on four messaging apps to get a one-on-one. He hands consistent performers more decisions until he sees bad ones and says a team that cannot absorb spiky talent is in a failure state.
Leaders should take the reins on high-stakes, one-way decisions. Francesco says good leaders delegate but are candid about when people need a decision made for them. Phil Carter adds that two-way door calls can go to junior people, with autonomy scaling with tenure and performance.
Suno launched as a Discord bot, then moved users to the web fast. Suno's first product was a crude Midjourney-inspired bot making 12.5 seconds of audio, chosen because Discord needed no UI or login and users could try, discuss and share in one place. When a basic web app launched in December 2023, 90% of traffic moved over within 5 days.
People spent hours with the bot despite poor audio. Cofounders pushed for the web app because users shouldn't have to recover a Discord password to use the product.
Halo AI runs its creator program as a Discord community. Each trial creator gets a private channel, with shared channels for wins, ideas and feedback, and creators chat daily in voice. When a video goes viral, an automation sends it to all creators to recreate the same day, and copies from 80 creators often get 2 to 3 million views each.
Trials pay $20 per video for five videos. Halo contrasts this with post-and-pay UGC programs that lose the shared learning.
Sellis was head of product here. A 2024 focus on friends playing multiplayer games moved users from about 10 active days a month to 12 or 13, driving some of the fastest growth since the pandemic, while Midjourney's boom was not visible in usage. Fixing bugs from Reddit threads won over critical gamers. It follows the same slow, roughly linear cohort penetration pattern as Snapchat Plus.
Discord grew fastest by removing friction for users it already had. In 2024 the team focused on being the best place for friends to play multiplayer games together, even though most such players already had Discord. Fixing bugs from Reddit threads and improving performance moved users from about 10 active days a month to 11, 12 or 13, driving some of the fastest growth since the pandemic.
Many users sat on the edge, using in-game voice or FaceTime instead. Midjourney's boom on Discord did not show up in the usage graph at all. Peter Sellis applied the same lesson at Snap, where Android performance work for existing users drove growth for years.
Define core value as a phrase, then turn each word into metrics. Discord's core value was talking and hanging out with friends before, during and after games, broken into metrics tied causally to daily return. It was used to judge roadmap items and to onboard new hires and data scientists, and the definitions went into Discord's growth playbooks.
Snap's version was the fastest way to share a moment with your best friends, which Sellis presented to every new Snap hire for seven years.
Find new bets by watching users bend the product to unplanned uses. Sellis cites Discord communities as users going through hoops to use the product in ways it was not designed for, one of two ways to spot expansion. The bolder way is backcasting from the founding idea with no tie to today's tech.
Subscriptions on messaging apps build slowly as cohorts form habits. Discord's subscription penetration followed the same slow, roughly linear cohort pattern as Snapchat Plus, because users must get used to daily use before they see the value. Sellis wonders if launching Snapchat Plus four years earlier would have eased pressure on ads.
"Time is your enemy with subscription products."
A shared mission plus named owners replaces most collaboration. As head of product, Sellis distilled the founders' vision into short repeatable phrases and gave each problem a directly responsible owner. PMs were pushed to make the calls the founders would have made a year or two earlier.
He argues collaboration makes a team as slow as its slowest node, and compares the model to a terrorist organization built on an ideology rather than a leader. Discord's chief people officer was "horrified and intrigued" by the framing.
Spiky talent like Nikita Bier is directed, not managed. Bier reported to Sellis for a little over a year after his startup was acquired, bringing strong engineers. Sellis says you don't manage him, you only seek to direct him at problems where he is expert.
Sellis once pinged Bier on four messaging apps to get a one-on-one. He hands consistent performers more decisions until he sees bad ones and says a team that cannot absorb spiky talent is in a failure state.
Growth stalled because non-gamers found it too complicated, so it redesigned mobile: replaced four-digit discriminators with usernames, added voice messages, rebuilt Hangouts video chat and clarified the profile tab. It monetizes with Nitro subscriptions and reward ads called Quests, keeping paywalls off the friend-invite loop.
Discord's founders took back UX from their least experienced people. They saw they had delegated the key product decisions to junior staff, so they flipped the pyramid and became the first and last mile on UX.
Mark Pincus uses this as an example of founders staying close to pixel-level detail for as long as they are the best player.
Its founders saw they had delegated UX, the key product decision, to their least experienced people. They flipped that pyramid and became the first and last mile of the product.
Suno launched as a Discord bot, then moved users to the web fast. Suno's first product was a crude Midjourney-inspired bot making 12.5 seconds of audio, chosen because Discord needed no UI or login and users could try, discuss and share in one place. When a basic web app launched in December 2023, 90% of traffic moved over within 5 days.
People spent hours with the bot despite poor audio. Cofounders pushed for the web app because users shouldn't have to recover a Discord password to use the product.
Suno launched its first bot there because it needed no UI or login, and users could try, discuss and share in one place. Its gaming culture encouraged long sessions, and about 90% of traffic left for the web app within 5 days of launch.
When funnel tweaks stopped working, Discord redesigned the product itself. Non-gamers kept saying they downloaded Discord after a friend's invite, found it too complicated and deleted it. The growth team concluded standard funnel work could not fix that and chose to redesign mobile, where most non-gamers joined.
Pandemic waves (Among Us, a WhatsApp outage, crypto communities) had brought crowds of newcomers in 2021. The project began as a hallway talk between Francesco Polizzi and early engineer Stan about making Discord less bloated. "We've optimized everything that's possible here. We need to change basically where we're optimizing from."
Start with two weeks inside every rival app, not a strategy doc. Francesco cleared his calendar and used Skype, Telegram, WhatsApp, Signal, Messenger and iMessage on throwaway accounts, testing every feature and dumping every gap into one master list. Grouping problems into teams came afterward.
Every problem fell into quality, usability or familiarity. The team held that quality and speed were not at odds and aimed to reach the first thing they could see and feel together, which told them what to do next.
Replacing four-digit discriminators let non-gamers actually find their friends. Usernames like name#1234 meant newcomers typed a friend's name without the number and failed to find them. After the switch to unique usernames, the share of outbound friend requests accepted rose by what Francesco called a pretty crazy amount.
The team also split DMs out of the messaging tab to feel familiar. It kept the gamer term "server" but made servers feel less foreign. No exact figure was given for the friend-request lift.
A one-question survey justified voice messages despite a rival's failure. A PM whose nephew loved voice messages pushed back on an anecdote that the feature had failed elsewhere. A quick survey of a subset of users found 70 to 80% sent voice messages weekly, and the shipped feature lifted overall communication by a single-digit percent.
Usage was higher in South America and the Middle East, which Francesco reads as a generational gap. Surveys work when they ask about one concrete existing behavior: "How often do you do this thing that already exists somewhere else?"
Rebuilding buggy video chat raised video minutes between friends. Mobile video froze when users zoomed, so the team rebuilt Hangouts, a months-long and technically hard project. It increased video minutes between friends.
Francesco frames it as hunting the nooks and crannies where bugs pile up in a fast-scaled product, with video as one of the most intimate ways to connect.
A clearer profile tab leaned into Gen Z's taste for expression. The redesign made the profile tab clearly about how others see you and made customization easier to reach. That foundation later supported paid animated profile headers, effects and the shop.
Map where new and power users clash before shipping big changes. Discord's rule was that the new audience should be a superset of existing users and nothing should alienate them, yet the team never pinned down where the groups would conflict. Forcing everyone to change usernames caused a large uproar that still draws online noise.
Francesco says the disconnect with core power users was discounted more than it should have been, and no count of lost users was given. Phil Carter's advice for such moments: wait a few weeks past the vocal first-week backlash, then compare it with the lasting effect on metrics.
Charge for power-user perks, never for what drives friends inviting friends. Discord's growth loop is friends inviting friends, so Nitro charges for perks like larger file uploads and custom emojis that never block a new user joining a server. Display ads fit poorly because most chat happens among two to four people with little inventory.
At Photomath, a rival capped scans and damaged its core loop, while Photomath monetized something else and reached about a 100x outcome by Francesco's estimate. Paywalled features must still be genuinely valuable to someone.
Reward ads called Quests fit gamers better than display ads. Brands show relevant game content and users earn orbs to spend on profile customization. Quests became Discord's main monetization push after Francesco left.
The shift followed the realization that the COVID audience expansion was less lasting than assumed and gamers remained the core of the business. Francesco offers the ads view as an outsider.
Audit screens leading into payment, since one bug can cost millions. Tapping the Nitro button on the profile sometimes failed to load the landing page or took about 30 seconds. Fixing it was described as worth millions of dollars.
The figure is anecdotal, with no exact number or control group.
Leaders should take the reins on high-stakes, one-way decisions. Francesco says good leaders delegate but are candid about when people need a decision made for them. Phil Carter adds that two-way door calls can go to junior people, with autonomy scaling with tenure and performance.
Halo AI runs its creator trials and community there, with a private channel per creator and shared channels for wins, ideas and feedback. Creators chat daily in voice, which keeps motivation high.
Halo AI runs its creator program as a Discord community. Each trial creator gets a private channel, with shared channels for wins, ideas and feedback, and creators chat daily in voice. When a video goes viral, an automation sends it to all creators to recreate the same day, and copies from 80 creators often get 2 to 3 million views each.
Trials pay $20 per video for five videos. Halo contrasts this with post-and-pay UGC programs that lose the shared learning.
Lovable's community started early on Discord and now has hundreds of thousands of members, run by community managers and an ambassador program.
Raiza pushed for a NotebookLM Discord server from day zero, despite colleagues suggesting Google Meet or Google groups. It grew to about 60,000 members and became the place for daily direct feedback.
Bought Gas for many millions after Gas ran lean as a cashflow business with no investors.
Serves communities, a different audience from teams doing work together, so Slack stays focused on groups working rather than competing head on.
Discord's roadmap closely matches Slack's, but its mission and audience differ, so the stack shows when copying a rival does not make sense.
Social & DatingA social network built around connecting people and sharing.
Social & DatingA photo and video sharing social app built around creators, Stories, Reels and feeds.
Social & DatingA short-form video app where creators post content and can earn through a revenue-sharing program.
Social & DatingA professional networking platform for jobs, hiring and career content.
Social & DatingA dating app with tiered subscriptions plus boosts, Super Likes and other in-app purchases.
Social & DatingThe company behind Snapchat, a camera-first messaging app popular with teens.
Social & DatingA camera-first messaging and social app built around close friends, Stories and AR lenses.
Social & DatingA social network for short public posts and real-time conversation.