Zynga

Copy the best of breed on your exact platform before innovating. Proven Better New was the engine of product management at Zynga and worked like a religion inside the company. When Sid Meier's social civilization game launched on Facebook, Zynga's product managers emailed within 10 minutes that it was dead on arrival because its first-time experience took too many clicks.

Even the most junior PM knew the best-of-breed onboarding on that platform. Proven counts only for the same platform, audience and experience, down to the pixel; a 1990s game is not proven for a mobile audience today. "We haven't earned the right to innovate on the camera until we are the world's leading PhD on the best mobile cameras that already exist."

Lenny's Podcast▶ 7:29

Better means a small polish 10 out of 10 existing users want. Words with Friends was Scrabble polished for mobile, plus one new idea: it sat on the Facebook graph, so friends were already there to play. It reached 14 million daily users while Scrabble itself did not.

Better can be as plain as free, no download or smoother mobile play, and power users notice it most. What a founder thinks is better is usually new. "Better is something that 10 out of 10 of your existing users would say [ __ ] yeah."

Lenny's Podcast▶ 8:30

Assume your new idea is wrong and keep four more ready. Pincus's rule of thumb is that instincts are right 95% of the time but ideas are right at best 25% of the time. So isolate the instinct, test many ideas around it, and fail for the right reason.

A team with four more new ideas queued runs the work differently. He says a Tribe-era team that tried more ideas would have changed its odds; Tribe had terrible D30 retention.

Lenny's Podcast▶ 4:05

Judge ambition by the end user's delight, not peers' respect. Pincus told Zynga product makers to burn their resume and win over the Indiana nurse who plays FarmVille. He calls founders' reluctance to copy a moral arbitrage open to anyone with less ego.

Pure copying gives a user no reason to switch, but one inch better can beat something wholly unfamiliar. The masters do it so well that nobody sees what it derives from.

Lenny's Podcast▶ 15:13

Draw Something won by copying Words with Friends' turn-based system. OMG Pop, on its last dollar after an all-new failure, copied the turn-based system of Words with Friends. Draw Something was the number one app for 60 days in 2012, and Zynga bought the company.

Pincus contrasts this with wildcat drilling from a blank page, such as Rovio making 45 unrelated games before Angry Birds.

Lenny's Podcast

Zynga built against day 365 retention, read through D1 and D30. Zynga tracked day 365 retention, which Pincus thinks no one else did, and he says the most valuable companies have the highest D365. Low D1 and D30 rule out a high D365, but high D30 can still lead to zero D365.

The question to ask is whether a user would still use this in a year, because users make that same judgment about whether to invest. Words with Friends outlasted FarmVille because it moved to mobile.

Lenny's Podcast▶ 44:00

Count friend round trips and build product to raise that number. Active Social Network counted round trips with a friend, such as a turn and a turn back or a gift and a gift back. Going from zero to one gave an 80% chance the player returned the next month, and at four there was an 80% chance of seeing them 22 of the next 30 days.

Lenny's Podcast▶ 46:10

Virality alone is a sinking speedboat; retention made Zynga's hits last. Eight of 10 large Zynga game launches became massive hits, many of them copies of competitors, and Pincus credits a clear mission and retention as the core metric rather than spam. "It wasn't that we were good at virality."

He adds that distribution is now weak, with average app installs per user per month at zero, so it must be built into the product from the start. His framework rests on his own hits, with no count of products that followed it and failed.

Lenny's Podcast▶ 40:08

Test launch marketing on your own game board before buying ads. For FarmVille's English countryside expansion, Zynga placed differently styled locked art on the game board, reaching 25 to 30 million daily players, with clicks leading to a coming soon page selling early access. It sold $19 million of early-access keys, against a planned $10 million ad budget.

The teasers measured which look and message drew clicks, gave direction on the product, and created revenue and scarcity at once. Pincus presents it as killing hope early by testing ideas cheaply.

Lenny's Podcast▶ 37:00

Start embarrassingly small, especially after a past success. After Tribe failed by trying to do all of social networking without picking one use case, Pincus started Zynga as a Facebook poker game at 41, a 1,000-foot ambition rather than 100,000. People thought it showed he had no dignity; it became a huge success.

Repeat founders can easily raise and hire against a big vision before product market fit. He later pulled the plug on his Earth project after four years and $25 million. "The more ambitious you are, the more humble you should be and the smaller place you should be willing to start."

Lenny's Podcast▶ 23:55

Founders should own pixel-level product decisions while they're the best player. At 50 employees, Pincus ran a two hour standup naming each person and what they would do that day. "Micromanagement is beautiful."

He points to Discord's founders taking the first and last mile on UX, and to Bezos's tech assistants as a way to pass judgment on.

Lenny's Podcast▶ 1:18:18