The hidden pattern behind successful products | Mark Pincus (Founder of Zynga)

An Overview of Mark Pincus's Proven Better New Product Framework

Mark Pincus, founder of Zynga, developed the 'Proven Better New' framework as a core principle for product management, designed to significantly increase the odds of successful product ideas. This framework helps founders and teams refine their concepts by leveraging human instincts and iterative testing, rather than relying solely on ambition or gut feelings.

The framework addresses common pitfalls in product development, such as founders stubbornly sticking with losing ideas. Pincus observes that while instincts about what might be good are often right, initial product ideas are wrong a majority of the time. A truly successful product, he notes, clearly signals its potential; if a product's success is in question, it's likely not an A-level idea.

A key aspect of this philosophy is defining ambition through the consumer's eyes, not by seeking industry awards or peer respect. Pincus emphasizes building products that genuinely win the hearts and minds of the target audience, drawing from his experience creating over a dozen successful consumer products, including those at Zynga, by focusing on what resonates with users.

If you're asking whether or not your product is an A, it's not an A.

Applying the Proven, Better, New Product Framework

Mark Pincus introduces the 'Proven, Better, New' framework, a core philosophy for product development that isolates innovation. This method challenges the instinct that new ideas are always right, stating that instincts are often correct, but ideas built upon them frequently fail. The framework aims to increase the odds of product success by guiding developers to test many ideas and fail for the right reasons.

The 'Proven' component emphasizes mastering existing, successful best practices before innovating. An example given is Sid Meier's social civilization game, which failed due to a poor First-Time User Experience (FTUE), despite his reputation. It neglected to copy the proven onboarding methods that even junior Zynga product managers understood. Products should become 'PhDs' in what already works, copying successful elements legally and tastefully.

The 'Better' stage involves small, incremental improvements that garner near-universal approval from existing users—ideally 100% saying 'Fuck yeah!'. These are enhancements that are statistically verifiable and often relate to polish, like making a product free or removing a download. Words with Friends, for instance, offered a highly polished mobile Scrabble experience, which constituted its 'Better' aspect.

Finally, 'New' refers to the novel idea that provides a compelling reason for people to download and try the product. This is the riskiest part, and the assumption should be that these ideas are likely to fail. Words with Friends' 'New' element was its social integration with the Facebook graph, allowing easy play with friends. By separating these components, teams can develop products with a higher probability of market success, ensuring that innovation isn't hampered by neglecting fundamental, proven elements.

Better is something that ten out of ten of your existing-- the existing users of that product would say, 'Fuck yeah!'

Innovation in product development often leverages 'proven' ideas with small, impactful 'better' improvements.

Product development can benefit significantly from a 'proven, better, new' framework. This approach involves identifying what is already working well in the market for a specific platform and audience, making it demonstrably better in a way that users would readily switch, and then optionally adding a truly novel feature. The key is to be extremely precise about what counts as 'proven,' ensuring it's relevant to the current context rather than broadly applying outdated concepts.

Many successful products, including the iPhone, Slack, and even Craigslist's gradual evolution, illustrate this strategy. The iPhone built on the proven concept of a music player, made it better, and introduced the new element of a touchscreen. Slack refined existing team communication, while Craigslist's creator spent two years carefully integrating photos into listings, understanding that even small improvements to beloved, established products are valued by users.

This strategy introduces a concept of 'moral arbitrage,' where copying successful existing work, though it can feel counter-intuitive for innovators, is a highly effective path. Consumers generally resist drastic change and prefer familiar behaviors with minor, but significant, improvements. Defining ambition through the consumer's experience rather than peer recognition helps overcome the ego-driven resistance to adopting and refining existing successful ideas.

Ultimately, the art lies in making these improvements so seamlessly that users don't even realize the product is derivative. People appreciate functionality they already understand, enhanced in a way that makes their experience easier or more enjoyable. Presenting something entirely new that users didn't know they wanted can be a harder sell than refining something they already love.

If you took something that she loves and you make it one inch better, she might love that more than if you showed her something she's never seen before and didn't wake up knowing that she wanted.

The Paradox of Starting Small for Ambitious Products

Truly ambitious and successful products often begin from surprisingly humble and "embarrassingly small" starting points. Founders who are too visionary or ambitious at the outset risk missing product-market fit by not starting simple enough. This counterintuitive approach suggests that restraint in initial ambition can lead to greater long-term success.

Mark Pincus illustrates this with his own career, recalling how Facebook began as a simple app for Harvard students. After an overly ambitious venture with Tribe failed, he launched Zynga Poker, a seemingly minor Facebook app, despite being a multi-time founder. This humble start was key to its massive success, highlighting that major achievements often stem from unassuming origins.

New founders possess an advantage over established ones because they lack the "rope to hang themselves" – the ability to raise significant funds for large visions before achieving product-market fit. This forces them to start smaller and more humbly. Startups are uniquely positioned to explore "little threads" that larger companies, constrained by high revenue expectations, cannot.

An example of this is Bolt.new, which started by toiling in obscurity on a niche technical interest. They eventually open-sourced their work and later combined their web stack virtual machine with an AI coding copilot, discovering a significant, differentiated product. This demonstrates how focusing on a small, specific passion can unexpectedly lead to a breakthrough.

The paradox is, the more ambitious you are, the more humble you should be and the smaller place you should be willing to start.

Kill Hope Before Hope Kills You

Mark Pincus advises founders to "kill hope before hope kills you" in product development, emphasizing the distinction between hope and belief. Hope is defined as confidence without basis, akin to a prayer, whereas belief is grounded in lived experience, product data, and user interaction. Many teams prolong efforts based on the hope that a future release will magically succeed.

Instead of launching a Minimum Viable Product (MVP) driven by hope, Pincus advocates for a Maximum Launchable Product (MLP). An MVP often relies on the word "viable," which he suggests is where hope originates, implying it "might make it." In contrast, an MLP is launched when the team has a firm belief, not just hope, that it will be a hit.

He notes that the best product makers, like Brian Chesky at Airbnb, launch products knowing they are hits rather than to discover if people like them. This approach means "collecting winnings" instead of making bets. The rapid development capabilities offered by AI can quickly lead to "viable" products in months instead of years, which Pincus warns is a dangerous drug if it encourages hope-driven launches rather than belief-backed ones.

Founders need the courage to tell their teams and investors when a product isn't a hit, even if it's a B+ with some traction. The ambition should be to relentlessly pursue a North Star product-market fit, rather than settling for something that merely seems viable.

Kill hope before hope kills you. There's a difference between belief and hope. Hope is confidence without basis.

Leveraging AI for Rapid Product Iteration and Learning

Entrepreneurs should use AI as a 'testing machine' or 'failure machine' to rapidly iterate through hundreds of ideas, rather than spending months developing a single product. The goal is to learn quickly, even by building something

wrong

initially to get signal and test core hypotheses.

Instead of building a perfect product, focus on creating something

build it completely wrong before we know it's the right product, so build it wrong before you know it's right.

Zynga's Success Stemmed from Social Game Design and Long-Term Retention Focus

Mark Pincus clarifies that Zynga's high hit rate, including major successes like Farmville and Cityville, wasn't primarily due to virality. Instead, the company differentiated itself by focusing on adding social dimensionality to games, a concept he calls a "bold beat" where new dimensions are added to existing ideas.

Zynga aimed to connect the world through games, enabling players to "Invest, Express, and Connect." This involved features like co-op play and gifting, allowing users to feel creative and improve relationships. For instance, middle-aged women enjoyed Farmville because it provided a social hobby where they could engage with friends and make new ones.

The second core driver was an obsessive focus on day 365 retention, a metric Zynga uniquely tracked as a consumer company. Pincus views high day 365 retention as a hallmark of the most valuable companies. While correlated with shorter-term metrics like D1 and D30, it requires a distinct mentality to build products that users will continue to invest in a year later, unlike many viral "sinking speedboats."

To operationalize this, Zynga developed the Active Social Network (ASN) metric, which measured reciprocal interactions between players, such as turn-taking or gifting. They found that increasing a player's ASN from zero to one drastically increased their likelihood of returning in the next month, and an ASN of four ensured frequent engagement over the following 30 days.

Viral based companies... they're sinking speedboats. They're trying to drive faster than they're sinking.

Reinventing Social: Latent Demand and the AI Cocktail Party

Many existing social media platforms, like Instagram and TikTok, have lost their "adrenaline" and now feel like "empty calories," causing users to quit them with a sense of pride, similar to quitting smoking. Despite this, there's a significant latent demand for social interaction, analogous to the massive gaming industry where few adults actively play yet it thrives. This suggests a ripe opportunity to reinvent social experiences.

The next generation of social platforms is expected to reintroduce "productivity" to the user experience, building upon the initial value propositions of early Facebook and LinkedIn, which offered significant social productivity. These new experiences will leverage agentic AI to help users stay connected and informed with people they care about, without the time-wasting elements that current platforms often induce.

A key concept for reinventing social is the "cocktail party," which naturally facilitates valuable "lead generation." This means connecting people for specific, productive outcomes. Historically, platforms like Napster provided "music file leads," Friendster enabled finding "dates," and LinkedIn offered professional connections. The challenge is to replicate this utility, where the value lies in meaningful connections rather than just engagement.

Current AI environments, like interacting with Claude or GPT, are akin to a "quiet, lonely cocktail party." The opportunity lies in transforming these spaces into "rowdy" and "socially productive" environments. By figuring out how to make AI-driven interactions genuinely social and productive, creators can uncover significant value and truly reinvent the social experience for the agentic AI age.

My challenge to your listeners is, figure out how to make it rowdy, figure out how to make that cocktail party social and socially productive, and you will probably find gold there.

Intellectual Honesty in Identifying and Actioning a 'B+' Product

If you are questioning whether your product is an 'A', it likely isn't. True 'A' products exhibit undeniable signals, like being addictive and showing strong metrics, much like the clear impact of GPT. There isn't a need to ask if it's an 'A' if it truly is one.

The first step is intellectual honesty: acknowledging when a product is merely 'B+'. The real power comes from knowing it's a 'B+' and then deciding what action to take. Options include killing it, using it as a learning experience, or attempting to iterate.

Mark Pincus shared his personal experience with 'dot Earth', a metaverse project he had worked on for 20 years, investing $25 million in one version. He ultimately pulled the plug on it because it wasn't achieving product-market fit and he was being too ambitious.

This difficult decision, though costly, proved beneficial. Within two weeks of ending the 'dot Earth' project, Pincus found himself more inspired with new ideas than he had been in the previous four years, highlighting the value of letting go of underperforming projects.

The power is knowing it's a B plus.

AI is Not Yet a True Distribution Platform for Consumer Products

Mark Pincus contends that AI, while an important technology, has not yet evolved into a new platform in the traditional sense. Unlike previous hardware platforms or interface platforms like Windows, browsers, or mobile operating systems that fostered discovery, current AI chat interfaces do not serve as platforms for other apps or developers. This puts us in a

halfway to a platform

state, hindering a new wave of consumer app discovery.

The current environment for consumer app discovery is dire. The average user installs zero new applications per month. This lack of discovery means that even with tens of thousands of new games launched, none achieved significant success, making consumer categories like social and games extremely difficult to invest in with high confidence.

The average app installs per user per month is zero.

Exploring Novel AI Distribution Strategies and Agentic Services

The current landscape presents significant distribution challenges for new products, making innovative approaches critical. One intriguing concept, popularized by Gary Tan, suggests that future AI tokens will effectively become free. This anticipated drop in the cost of AI compute could unlock a wave of new consumer services that were previously uneconomical or impossible to build, distinct from past internet businesses where underlying costs remained high.

Another frontier involves agentic services that can skillfully broker human relationships. Imagine an AI "social secretary" with an "intelligent membrane of trust" that understands the context of both parties. This agent could facilitate social interactions, appointments, or even dating, by sharing only necessary information and navigating social complexities to avoid awkwardness or perceived disinterest.

A prime example of an "amazing service" is a 24/7 AI travel agent. Unlike human agents who couldn't sustain a business given low commissions and unwillingness to pay, an AI could offer continuous, personalized support. This includes booking, understanding travel context, and proactively rebooking flights or managing logistics during a trip. This represents a significant shift from the internet's early focus on "better deals" to delivering truly transformative services. It's envisioned that LLM providers like OpenAI might create dedicated consumer platforms for such agentic applications.

The internet has grown through us getting a better deal. But I think the next area of growth probably will include deals, but I think it's going to be about getting amazing services.

Empowering CEOs: Stay Close to the Metal and Micromanage

Founders and CEOs should maintain a deep involvement with the product, a concept known as staying 'close to the metal.' This means being in the trenches, working with primary data, and understanding the minute details, rather than delegating all product decisions. The best product makers, even at a CEO level, should be on the field, similar to how Steve Jobs was deeply involved in design choices or how Jeff Bezos and Zuck would spend dedicated time with their teams on key products.

This philosophy counters the 'expert witness syndrome' where individual contributors are closest to the right answer but are furthest from making decisions. Many founders emerge from this frustration, seeking to retain decision-making power. Mark Pincus advocates for an 'inverted pyramid' approach, where founders are the first and last mile for product decisions, especially UX, ensuring that their time is spent on minute details that enhance the user experience.

A core tenet of this approach is embracing 'micromanagement,' particularly for early-stage companies. Mark recalls micromanaging Zynga with 50 employees, using detailed stand-ups to track individual progress. The idea is for leaders to be in the room and deeply involved as much as possible, only delegating when it's physically impossible to be present. This is less contrarian now, being recognized as an effective practice for driving results.

To scale this hands-on approach, leaders can use 'non-scalable ideas that actually scale.' This includes creating a 'teaching hospital' environment where a leader's passion and insights are directly shared with many people by having them in the room during critical product meetings. Another strategy is to utilize 'tech assistants' to help spread the leader's vision and approach throughout the organization, effectively passing on the 'vampire blood' of their mission.

Micromanagement is beautiful, you should micromanage as long as you can.

A CEO's Foremost Duty Is To Be Right

Mark Pincus contends that the number one job of a CEO, a principle he learned from Jeff Bezos, is to "be right." This means making correct product and strategy decisions is the most critical responsibility.

Pincus argues that getting the strategy right is more important than flawless execution, inspiring leadership, or meticulous management. He uses the analogy that being in the "right body of water" matters significantly more than having the "right boat," as an excellent boat in a dead lake bed won't go anywhere.

This philosophy also guides Pincus when evaluating individuals. He actively looks for people who have a proven track record of being right about something, considering it the most valuable trait on a resume and within a team, rather than focusing on their personal style or approach.

Being in the right body of water matters more than the right boat.

Parenting in the AI Era: Fostering Critical Thinking and Generative Habits

Mark Pincus emphasizes a parenting philosophy of meeting children where they are, treating them as humans, and engaging them at their altitude. He applies this by nurturing their natural curiosity, as exemplified by teaching his twin daughters "daddy math" during the pandemic, which covered material through eighth grade without him or them realizing it.

Pincus believes the hundred-year cycle of mass-produced, knowledge-based education is ending as AI changes the nature of work. He focuses on developing critical thinking and usefulness in his children over traditional college paths. For instance, his neurodivergent daughter, Carmen, created a brand "Comfy Fancy" and a support group "NeuroSparkly" for middle schoolers, transforming a perceived deficit into a way to connect and help others.

He encourages his children to ask better questions and understand underlying agendas, rather than just knowing answers. In their online lives, Pincus promotes generative habits—creating and putting new things into the world—over passive consumption of content and experiences. He views this as crucial for them to get something meaningful back from their interactions.

Pincus also shares fundamental life philosophies with his older daughters in a Google Doc, including the principles that "nothing's personal" and to "not be a victim." These teachings aim to equip them with resilience and an understanding that their reaction to events, not the events themselves, defines them.

I don't care if you go to college, what I care about is that you develop critical thinking and you find a way to be useful to people.

Mark Pincus's Vision for Building Internet Treasures

Mark Pincus shares his core ambition: to create "internet treasures." These are defined as services so indispensable that people cannot remember a time before them or imagine life without them. He believes this represents the highest aspiration for product makers, allowing them to offer significant value to the world.

Pincus emphasizes the monumental opportunity for product makers in this era to construct these digital "skyscrapers." He envisions them becoming so fundamental that future generations will find it difficult to believe anyone ever lived without them, with a friend suggesting they could one day be featured in the Smithsonian.

This ambition to build internet treasures is also the driving force behind his book, "Life at the Speed of Play." Pincus intends the book to be a playbook, sharing his philosophies and ideas with the hope that product makers will adopt, adapt, and expand upon them, thereby advancing the craft of product creation.

to build an internet treasure, which is a service we can't remember life before or imagine life without.