Mark Pincus
Founder of Zynga, author of Life at the Speed of Play
Where to hear and read about Mark Pincus1 episode or test
Mark Pincus founded Zynga, which launched hits like FarmVille and Words with Friends, and wrote Life at the Speed of Play. The lesson most worth taking is to copy what is proven on your platform, make it a small step better, add one new idea, and test many cheap ideas so the losers die early.
The more ambitious you are, the more humble you should be and the smaller place you should be willing to start.
The hidden pattern behind successful products | Mark Pincus (FarmVille, Words with Friends, & more)
Mark Pincus: the story and the lessons
1Pincus's Proven, Better, New framework for product ideas
Pincus trusts his instincts far more than his ideas. His rule of thumb is that instincts are right 95% of the time and ideas are right at best 25% of the time.
So he isolates the instinct, then tests many ideas around it, so a product fails for the right reason and not the wrong one.
He copies what is proven on the exact platform first. Proven counts only for the same platform, audience and experience, down to the pixel. Zynga's product managers called Sid Meier's Facebook game dead on arrival within 10 minutes because its first-time experience skipped the best onboarding pattern on that platform.
Founders misuse "proven" to defend a wrong idea, and a 1990s game is not proven for a mobile audience today. His line: "Get your PhD in proven first."
Better means 10 out of 10 existing users say yes. Better is small, provable polish, such as free or no download. Words with Friends was Scrabble polished for mobile, with friends already on the Facebook graph, and it reached 14 million daily users.
What a founder thinks is better is usually new, and founders are usually too ambitious about it. Power users notice polish most.
He assumes the new idea will fail and keeps four more ready. The new step is the one wrinkle nobody has tried. Pincus expects it to be wrong most of the time, so a team should hold other new ideas to test.
He says a team with four more ideas runs its work differently. His Tribe had terrible D30 retention, and he now sees that his Tribes idea was Reddit.
He calls copying a moral arbitrage for founders with less ego. School taught that copying is cheating, so few ambitious founders will do it. Pincus told Zynga's makers to burn their resumes and win over the Indiana nurse who plays FarmVille, not impress peers.
The best proven idea is often buried in someone else's product. Freeloader was built on offline browsing hidden in Netscape and Internet Explorer. OMG Pop copied the turn-based system of Words with Friends in Draw Something, which was the number one app for 60 days in 2012, and Zynga bought the company.
2Why Pincus says to start small and kill hope early
He starts embarrassingly small, especially after past success. Zynga began as a Facebook poker game, which Pincus calls a 1,000-foot ambition, when he was 41. Tribe tried to do everything and never picked one use case.
Repeat founders have too much rope: it is easy for them to raise money and hire against a big vision before product market fit.
Pincus separates belief from hope and launches only hits. He defines hope as "confidence without basis", while belief rests on experience with the product and the data. He says the best makers are collecting winnings, not making bets, and cites Brian Chesky as knowing he has a hit before launch.
He wants the word "viable" killed in minimum viable product, since a viable product feeds hope. His line: "Kill hope before hope kills you."
He shut down his Earth project after four years and $25 million. Earth was a web game engine for his version of the metaverse. He pulled the plug for the fourth time and felt more inspired about ideas in the next two weeks than in four years.
He says he "went native, not in the good way", so keen to prove he could build it that he never reached product market fit. His test: if you ask whether the product is an A, it is not an A.
He wants AI used to test a hundred ideas, not build one. A viable product now takes about 3 months instead of 3 years, which he calls dangerous because it feeds hope. He would rather teams build it wrong in a day or a week to get signal.
He expected people to build testing machines that run more ideas in a week than an industry tests in a year. Instead, most use AI to build one idea in three months.
Pincus tested launch marketing inside the live game. For the FarmVille English countryside expansion, he put locked art variants on the game board in front of 25 to 30 million daily players instead of spending a planned $10 million ad budget. Clicks showed the best look and message, and early-access keys sold $19 million.
The test gave product direction and revenue at once.
3What made Zynga's games last
Zynga's hit rate came from a clear mission, not virality. Eight of 10 large launches became massive hits despite many copycats. The mission was connecting the world through games, adding a social "bold beat" to old games.
FarmVille and CityVille were the biggest hits, but Words with Friends and Poker lasted longer because they moved to mobile. Pincus says the games' feed takeover also bugged friends who didn't play.
He tracked day 365 retention when he thinks no one else did. Low D1 and D30 predict low D365, but a high D30 can still end at zero by day 365. Pincus calls viral companies "sinking speedboats": plug the hole and build a better boat instead of bailing faster.
He asks whether a user would still use the product in a year.
Zynga built a social loop metric called Active Social Network. ASN counted round trips with a friend, such as a turn taken and returned or a gift and a gift back. Going from zero to one gave an 80% chance of seeing the player the next month, and at four there was an 80% chance of seeing them 22 of the next 30 days.
Product work was then aimed at raising that number.
He found latent demand in a category that looked dead. In 2007 gaming was a $23 billion business, yet Pincus knew no one who played and found playing with others like crawling across broken glass. He made games free, familiar, three clicks and five to 15 minutes.
Gaming is now a $280 billion industry, and he thinks adults still mostly don't play.
He says distribution must be built into the product now. About 40,000 games launched in the app store last year and none became a top 10 hit or held the top 25 or 50. He calls relying on virality or word of mouth a hope strategy.
He suggests starting with prosumers and power users who pay up front, and designing today for tokens that will be nearly free in a year or two, an idea he credits to Gary Tan.
4How Pincus leads, hires and stays in the details
He makes key people CEOs of a hill. He gives a person a hill to take and real operating control, and asks for a plan and budget. A CEO with those does not come back with questions.
He says "All of management is just how do we get people to do the right thing when we're not in the room." He hires "frustrated expert witnesses", know-it-alls who want to prove they were right.
Pincus says micromanagement is beautiful while you are the best player. At up to 50 employees he ran a two-hour daily standup that went name by name through yesterday's and today's tasks, and the team was much more effective.
He cites Discord's founders, who flipped the pyramid after handing UX to their least experienced people. He says delegation should come last.
He puts people in the room and picks one as a shadow. He fills product management meetings with as many people as possible so his approach rubs off. He picks someone from the ranks as tech assistant for six to twelve months, a "mini me" who gets test projects and then a much bigger role elsewhere.
He points out that everyone on Andy Jassy's S-staff at Amazon had been tech assistant to Bezos.
He hires for a record of being right, taking Bezos's view. Pincus says the number one job of a CEO is to be right about product and strategy. Being in the right body of water matters more than the right boat.
He will take misfits who are right and intellectually honest, because people who are right "put balls in the net".
He admits his transparency burned out teams at Zynga and since. Pincus leaned toward being transparent "to a fault". The critique of working for him is that it is like third grade soccer, with a new idea every Monday.
He concedes the critique is fair and says he is still learning.
5Pincus's purpose and his book
He wants to build an internet treasure. He only put his why into words when he started Zynga at 41. An internet treasure is a service people can't remember life before.
Answering those who call him post-economic, he says "I haven't done my thing yet."
Life at the Speed of Play is a short playbook to borrow. The book took five years to write, and for three of the four writing years it was titled Proven Better New. He dropped the title as too dry.
He wrote it bite-sized because he has trouble reading many books, and he sees product makers as one conversation, all learning as they go.
The hidden pattern behind successful products | Mark Pincus (FarmVille, Words with Friends, & more)
Mark Pincus explains his Proven, Better, New framework for product ideas, why starting small beats starting ambitious, and why founders should kill hope early by testing many cheap ideas. At Zynga, eight of 10 major game launches became massive hits, and he credits retention and an obsessive product process over virality.
Pincus trusts his instincts far more than his ideas. His rule of thumb is that instincts are right 95% of the time and ideas are right at best 25% of the time.
So he isolates the instinct, then tests many ideas around it, so a product fails for the right reason and not the wrong one.
He copies what is proven on the exact platform first. Proven counts only for the same platform, audience and experience, down to the pixel. Zynga's product managers called Sid Meier's Facebook game dead on arrival within 10 minutes because its first-time experience skipped the best onboarding pattern on that platform.
Founders misuse "proven" to defend a wrong idea, and a 1990s game is not proven for a mobile audience today. His line: "Get your PhD in proven first."
Better means 10 out of 10 existing users say yes. Better is small, provable polish, such as free or no download. Words with Friends was Scrabble polished for mobile, with friends already on the Facebook graph, and it reached 14 million daily users.
What a founder thinks is better is usually new, and founders are usually too ambitious about it. Power users notice polish most.
He assumes the new idea will fail and keeps four more ready. The new step is the one wrinkle nobody has tried. Pincus expects it to be wrong most of the time, so a team should hold other new ideas to test.
He says a team with four more ideas runs its work differently. His Tribe had terrible D30 retention, and he now sees that his Tribes idea was Reddit.
He calls copying a moral arbitrage for founders with less ego. School taught that copying is cheating, so few ambitious founders will do it. Pincus told Zynga's makers to burn their resumes and win over the Indiana nurse who plays FarmVille, not impress peers.
The best proven idea is often buried in someone else's product. Freeloader was built on offline browsing hidden in Netscape and Internet Explorer. OMG Pop copied the turn-based system of Words with Friends in Draw Something, which was the number one app for 60 days in 2012, and Zynga bought the company.
He starts embarrassingly small, especially after past success. Zynga began as a Facebook poker game, which Pincus calls a 1,000-foot ambition, when he was 41. Tribe tried to do everything and never picked one use case.
Repeat founders have too much rope: it is easy for them to raise money and hire against a big vision before product market fit.
Pincus separates belief from hope and launches only hits. He defines hope as "confidence without basis", while belief rests on experience with the product and the data. He says the best makers are collecting winnings, not making bets, and cites Brian Chesky as knowing he has a hit before launch.
He wants the word "viable" killed in minimum viable product, since a viable product feeds hope. His line: "Kill hope before hope kills you."
He shut down his Earth project after four years and $25 million. Earth was a web game engine for his version of the metaverse. He pulled the plug for the fourth time and felt more inspired about ideas in the next two weeks than in four years.
He says he "went native, not in the good way", so keen to prove he could build it that he never reached product market fit. His test: if you ask whether the product is an A, it is not an A.
He wants AI used to test a hundred ideas, not build one. A viable product now takes about 3 months instead of 3 years, which he calls dangerous because it feeds hope. He would rather teams build it wrong in a day or a week to get signal.
He expected people to build testing machines that run more ideas in a week than an industry tests in a year. Instead, most use AI to build one idea in three months.
Pincus tested launch marketing inside the live game. For the FarmVille English countryside expansion, he put locked art variants on the game board in front of 25 to 30 million daily players instead of spending a planned $10 million ad budget. Clicks showed the best look and message, and early-access keys sold $19 million.
The test gave product direction and revenue at once.
Zynga's hit rate came from a clear mission, not virality. Eight of 10 large launches became massive hits despite many copycats. The mission was connecting the world through games, adding a social "bold beat" to old games.
FarmVille and CityVille were the biggest hits, but Words with Friends and Poker lasted longer because they moved to mobile. Pincus says the games' feed takeover also bugged friends who didn't play.
He tracked day 365 retention when he thinks no one else did. Low D1 and D30 predict low D365, but a high D30 can still end at zero by day 365. Pincus calls viral companies "sinking speedboats": plug the hole and build a better boat instead of bailing faster.
He asks whether a user would still use the product in a year.
Zynga built a social loop metric called Active Social Network. ASN counted round trips with a friend, such as a turn taken and returned or a gift and a gift back. Going from zero to one gave an 80% chance of seeing the player the next month, and at four there was an 80% chance of seeing them 22 of the next 30 days.
Product work was then aimed at raising that number.
He found latent demand in a category that looked dead. In 2007 gaming was a $23 billion business, yet Pincus knew no one who played and found playing with others like crawling across broken glass. He made games free, familiar, three clicks and five to 15 minutes.
Gaming is now a $280 billion industry, and he thinks adults still mostly don't play.
He says distribution must be built into the product now. About 40,000 games launched in the app store last year and none became a top 10 hit or held the top 25 or 50. He calls relying on virality or word of mouth a hope strategy.
He suggests starting with prosumers and power users who pay up front, and designing today for tokens that will be nearly free in a year or two, an idea he credits to Gary Tan.
He makes key people CEOs of a hill. He gives a person a hill to take and real operating control, and asks for a plan and budget. A CEO with those does not come back with questions.
He says "All of management is just how do we get people to do the right thing when we're not in the room." He hires "frustrated expert witnesses", know-it-alls who want to prove they were right.
Pincus says micromanagement is beautiful while you are the best player. At up to 50 employees he ran a two-hour daily standup that went name by name through yesterday's and today's tasks, and the team was much more effective.
He cites Discord's founders, who flipped the pyramid after handing UX to their least experienced people. He says delegation should come last.
He puts people in the room and picks one as a shadow. He fills product management meetings with as many people as possible so his approach rubs off. He picks someone from the ranks as tech assistant for six to twelve months, a "mini me" who gets test projects and then a much bigger role elsewhere.
He points out that everyone on Andy Jassy's S-staff at Amazon had been tech assistant to Bezos.
He hires for a record of being right, taking Bezos's view. Pincus says the number one job of a CEO is to be right about product and strategy. Being in the right body of water matters more than the right boat.
He will take misfits who are right and intellectually honest, because people who are right "put balls in the net".
He admits his transparency burned out teams at Zynga and since. Pincus leaned toward being transparent "to a fault". The critique of working for him is that it is like third grade soccer, with a new idea every Monday.
He concedes the critique is fair and says he is still learning.
He wants to build an internet treasure. He only put his why into words when he started Zynga at 41. An internet treasure is a service people can't remember life before.
Answering those who call him post-economic, he says "I haven't done my thing yet."
Life at the Speed of Play is a short playbook to borrow. The book took five years to write, and for three of the four writing years it was titled Proven Better New. He dropped the title as too dry.
He wrote it bite-sized because he has trouble reading many books, and he sees product makers as one conversation, all learning as they go.