Mark Pincus — unpacks his \"Proven, Better, New\" framework

In one sentence

The "Proven, Better, New" framework is a scientific method for product development that isolates innovation to specific variables while keeping the core, validated mechanics of a product untouched.

Why read this (and who it's for)

This summary is for product managers, founders, and engineers who struggle with "feature creep" or "innovation paralysis"—the tendency to change everything at once and lose the core value proposition of a product. By internalizing this framework, you will be able to strip away the ego of "originality" and focus on what actually moves the needle for users, allowing you to launch faster and iterate with scientific precision.

The big idea

Mark Pincus argues that most product failures stem from a lack of discipline in innovation. Founders often believe that to be "innovative," they must reinvent the wheel, but the most successful products are often those that master the "proven" mechanics of a category and only innovate on the "new" elements.

The core logic is based on instincts versus ideas. Pincus asserts that our instincts (the gut feeling that something is broken or a better way exists) are often right, while our specific ideas for how to fix it are usually wrong. Therefore, the goal of a product maker is to be a scientist: isolate the variables of a product so you can test your ideas without breaking the parts of the product that already work.

Key ideas

Instincts vs. Ideas

• What it is — Your gut feeling (instinct) that a problem exists is your most valuable asset; the specific solution you brainstorm (the idea) is almost always a starting point that will need to be discarded.
• Why it matters — If you attach your identity to a specific idea, you become defensive when that idea fails. If you attach yourself to the instinct that a problem needs solving, you remain open to pivoting as the data comes in.
• In the book — Pincus notes that we often have the "idea" for a dating app or a specific feature, but the real value is the instinct that "something was broken there" or "there was a better way."

The Proven, Better, New Framework

• What it is — A framework to categorize product elements into three buckets: Proven (the core mechanics that work), Better (the obvious improvements), and New (the novel, risky innovations).
• Why it matters — It allows you to "box in" your innovation. By identifying what is "Proven," you ensure you don't break the user's mental model. By identifying what is "New," you know exactly where the risk lies.
• In the book — Pincus uses the example of Threads. He argues that because it is a "fast follow" of a successful product, the "Proven" parts are the ones that already work for users on that platform. You should copy those legally and artfully, without changing a single thing.

The Art of Copying

• What it is — The best masters of the craft are the best at copying; they have the least ego regarding the "originality" of the core mechanic.
• Why it matters — Amateur product makers feel the need to change everything, which leads to "false negative" signals—the product fails not because the idea was bad, but because the foundational mechanics were broken.
• In the book — Pincuc points out that in 2004, there were three exact copies of LinkedIn in Chinese. The "Proven" part was the game mechanic and the demand; the goal is to stay as close to that as possible.

Defining "Better" vs. "New"

• What it is — "Better" is what 10 out of 10 existing users would say is an improvement (e.g., "it's now free" or "no download"). "New" is the novel idea that will likely fail but is the reason people try the product in the first place.
• Why it matters — Most people confuse "New" with "Better." If you think a feature is "Better" but it's actually just "New," you are guessing. "Better" should be obvious and immediate.
• In the book — Pincus uses the example of the board game Risk. A company changed the wooden pieces to plastic and added Roman numerals. This was "Proven Worse" and "Bad New." It wasn't "Better" because it didn't solve a user pain point; it was just a change for the sake of change.

The "Testing and Failure" Machine

• What it is — Instead of trying to build the "right" product, build a culture of building the "wrong" product quickly.
• Why it matters — Speed of engagement and learning is the only way to survive in a competitive market. If you aren't embarrassed by your first launch, you aren't launching early enough.
• In the book — Pincus advocates for testing more ideas in a week than your competitors test in a year. This means testing the "top of the funnel" (e.g., a landing page or ad link) before building the actual product.

Frameworks & tools

The Proven, Better, New Audit

Use this to categorize every feature or component of your current product:

1. Proven: What is the core "game mechanic" or "product mechanic"? What do users already do here? (Rule: Do not change these).
2. Better: What would 10/10 users agree is an improvement? (e.g., faster, cheaper, easier).
3. New: What is the novel "Back of the Box" feature? (This is where you allow yourself to experiment and fail).

The "Speed of Play" Testing Protocol

1. Identify the Instinct: What is the specific "broken" thing you see in the market?
2. Build the "Wrong" Version: Create a landing page or a simple ad link to see if anyone clicks.
3. Measure the Signal: Do not build the full product until you have a signal of interest.
4. Iterate Weekly: Give yourself and your team permission to change the "New" elements every Monday.

Memorable examples & stories

• The LinkedIn Copies: In 2004, Pincus saw three exact copies of LinkedIn in Chinese. The lesson: The "Proven" mechanic is the most important thing to replicate; the "New" stuff is secondary.
• The Risk Board Game: A company tried to "improve" Risk by changing pieces to plastic and adding Roman numerals. The lesson: If it isn't "Better" (solves a user need), it's just "Bad New."
• The "Internet Winter": Pincus and Hoffman recall the period when the internet was considered "concluded" (Yahoo era). The lesson: While everyone else was looking at the "Enterprise" shift, the "Consumer" revolution was just beginning.

Key quotes

• "The best product makers... are the best at copying. And they have the least ego around it."
• "What you believe is better is called new. What's actually better is what 10 out of 10 of the existing users of that product would say, 'Yeah.'"
• "Build a culture of building it wrong."
• "If you're not embarrassed, you know you're not launching early enough."

How to apply it

This week

• Audit your current project: Take a piece of paper and divide it into three columns: Proven, Better, New. Be honest—if a feature is just "New" but doesn't actually make the product "Better," consider cutting it.
• Identify one "Broken" Instinct: Write down one thing in your industry that feels wrong or inefficient. Do not write a solution yet; just name the problem.
• The "Embarrassment" Test: Look at your current MVP. If you aren't slightly embarrassed by how simple it is, you are likely over-engineering and moving too slowly.

Ongoing

• The Monday Pivot: Give your team permission to change one "New" element of the product every Monday.
• The 10/10 Rule: Before adding a "Better" feature, ask: "Would 10 out of 10 users say 'Yeah' to this?" If the answer is "Maybe," it's a "New" feature, not a "Better" one.

Common pitfalls

• The "Innovation" Trap: Thinking that "New" equals "Good." Pincus warns that "New" is often where products fail.
• The "Viable" Trap: Pincus argues that "Viable" is a weak word. You don't want a "viable" product; you want conviction.
• The "Pattern" Trap: Relying too much on what worked before (patterns) without looking for the "pattern interrupt"—the discontinuous change that leads to the best outcomes.

Connects to

• The Lean Startup (Eric Ries): Pincus mentions Ries but argues that "Viable" is a weak word compared to "Conviction."
• The "Internet Winter" Context: Connects to the history of the Dot-com bubble and the subsequent shift to Enterprise/SaaS.

If you remember only three things

• Instincts over Ideas: Trust your gut that a problem exists; be humble about the specific solution you propose.
• The Proven, Better, New Framework: Isolate your innovation. Keep the "Proven" mechanics stable while testing "New" ideas.
• Build to Fail: Build a "testing and failure machine" to get to the truth of the market faster than your competitors.

Internalize it

• Reflection questions
  • Which part of my current product is "Proven" and which part is "New"? Am I spending too much time on the "New" and neglecting the "Proven"?
  • If I had to launch a version of my product today that I was "embarrassed" to show people, what would it look like?
  • What is the "broken" thing I see in my industry that everyone else is ignoring?
• Flashcards
  • Q: What is the difference between "Better" and "New" in Pincus's framework?
  • A: "Better" is what 10/10 existing users would agree is an improvement (e.g., free, no download). "New" is the novel idea that is the reason people try the product, but is more likely to fail.
  • Q: Why is "Instinct" more important than "Idea"?
  • A: Because your instinct (that something is broken) is often right, but your specific idea for how to fix it is usually wrong.
  • Q: What does it mean to "build a culture of building it wrong"?
  • A: It means prioritizing speed of engagement and learning over a polished, "viable" product that takes too long to launch.
  • Q: How should you handle "Proven" mechanics when copying a competitor?
  • A: Stay as close to the original as possible without doing anything illegal; the goal is that the end user doesn't feel like it's a blatant copy.
  • Q: What is the "Embarrassment" rule?
  • A: If you aren't embarrassed by your launch, you aren't launching early enough to learn from the market.

Fidelity note

This summary is based strictly on the transcript provided. I am confident in the "Proven, Better, New" framework as it is the central pillar of the conversation. The section on "AI as a platform" is a nuanced discussion between Pincus and Hoffman; Pincus argues that while AI has "platform technology," it is not yet a "consumer platform" in the way Google was for the web. The "Risk" board game example is used specifically to illustrate the difference between "Better" and "New."