Proven Better New
Master the proven, add a verifiable better, gamble one risky new — to stack the odds a product wins
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 4
- Confidence
- 95%
A product-management engine for turning a raw instinct into a product with far better market odds. You separate what you are NOT innovating on (the proven, which you copy perfectly), from a small, statistically-demonstrable improvement (better, that 10 of 10 existing users would agree with), from the single risky new idea that gives people a reason to try it. Because the new idea will probably fail, you line up several to test rather than betting everything on one.
Origin
Developed by Mark Pincus at Zynga, where he says it became 'like a religion' and the core engine of product management. He credits the underlying 'moral arbitrage' framing of copying to Peter Thiel's contrarian thinking.
Core principles
- 01Your instincts are right ~95% of the time; your ideas are wrong ~75% of the time — isolate the instinct, distrust the idea
- 02Copy the proven legally and with taste; you haven't earned the right to innovate until you are the world's PhD on what already works
- 03'Better' is small, polished, statistically provable, and something 10/10 existing users would endorse — not you
- 04What you think is 'better' is usually actually 'new' — and new probably fails, so keep four more new ideas ready to test
- 05'Proven' is platform-, audience-, and experience-specific; a hit from another platform or era does not count
How to run it
- 1
Isolate your innovation zone
Identify the gut-level instinct you actually believe in and strip away everything you are not innovating on. Take all the proven elements off the table first so you can see exactly where the genuine bet sits.
Pro tip Think at the pixel level: name the specific platform, audience, and experience your instinct applies to.
- 2
Get your PhD in proven and copy it
Study the best-of-breed version of every element you are NOT innovating on — the icon, the camera behavior, the onboarding flow — and copy it perfectly, legally, and with taste. Be a master of the proven before touching anything new.
Pro tip For a first-time-user experience, clone the platform leader's flow click-for-click before adding anything.
Watch out Founders misuse 'proven' to justify a wrong idea by pointing to something popular on a different platform or decade. If it isn't proven on THIS platform for THIS audience, it does not count.
- 3
Add a 'better' that 10 out of 10 users would endorse
Layer in a small, polished, verifiable improvement — free, no download, faster, more accessible — that you can both show statistically and get every existing user to say yes to. Better is usually incremental and most visible to power users.
Pro tip If your 'better' makes only some users say yes, it's actually a 'new' idea in disguise — reclassify it.
- 4
Choose one 'new' and prepare backups
Pick the single novel 'back-of-the-box' idea that gives someone a reason to download and try the product. Accept it will probably fail, so have four more new ideas queued and look around the market for adjacent 'heat' you can test.
Pro tip When your metrics (e.g. D30 retention) are sinking, don't stoically stick with one new idea — cycle to the next queued bet.
Watch out Being too ambitious on the 'new' is the most common failure; keep the novel surface area small.
In the wild
The most revered game designer launched a social game, but Zynga PMs saw within 10 minutes that his first-time-user experience took too many clicks. He innovated on game design but failed to copy the proven onboarding flow.
→ 'Dead on arrival' — his great game design was never seen because he skipped the proven onboarding, while junior Zynga PMs already knew the best-of-breed approach.
OMG Pop first tried something totally innovative that no one had seen — it failed completely and left them on their last dollar. Desperate for a hit, they ran Proven Better New: copied a proven game and perfectly cloned Words with Friends' turn-based system.
→ Draw Something became the #1 app in the store for ~60 days in 2012 and Zynga acquired them.
Scrabble was the proven game; the 'better' was intense mobile polish; the 'new' was attaching it to the Facebook social graph so your friends were already there to play.
→ 14 million daily active users — a massive hit where plain Scrabble was not.
Common mistakes
Using 'proven' to launder a wrong idea
Founders point at a game popular in the 90s or on another platform and claim they're doing 'proven,' when proven only counts on the same platform, audience, and experience — this self-deception hides that they're actually shipping an unproven new idea.
Over-innovating on the 'new'
Treating your own untested idea as 'better' and building the whole product around it. New ideas fail ~75% of the time, so a product that innovates on many surfaces at once has terrible odds; the discipline is to keep the novel part tiny and copy everything else.
Is it for you?
Best for
Consumer product founders and PMs with a strong gut instinct who want to systematically raise the odds their next product succeeds
Not ideal for
Deep-tech or research projects where the entire value IS an unprecedented invention with no proven analog to copy
From the transcript
“your instincts are right 95% of the time. Your ideas are wrong 75% or at best right 25% of the time.”
“be a master of the proven first. Get your PhD in proven first.”
“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”
“let's take all the proven off the table”
“We usually are too ambitious on the new”
“It's dead on arrival because his Fatoule, his first time user experience, was so many clicks and so bad that no one was ever going…”
“they perfectly copied our turn-based system from Words with Friends and they made a massive hit.”
From the episode
The hidden pattern behind successful products
Mark Pincus (founder of Zynga)