Andreessen's Media Playbook: Talk Like You're at Dinner with Friends

Source: a16z | Published: 2026-06-19T14:40:13Z

Traditional media training tells you "never be interesting," but in the new media era, brand is personal — founders must discuss the world like Alex Karp, not pitch their company.

Andreessen said the advice hit him like a bolt of lightning — so obvious that no other media coach would ever give it to you. Zeldon's logic was simple: if you're going on stage to talk about something you don't know cold, you shouldn't be there in the first place. And if you do know the topic inside out, you should be able to make it just as vivid and compelling as when you're chatting with a friend.

The Golden Rule of Old Media: Never Be Interesting

The core objective of traditional media training was to minimize controversy. CEOs would walk out of an interview, and their proudest achievement was "I didn't make any news." Ben Horowitz boiled it down to one line: All news is bad news.

The training process itself tells you everything — you're parked in front of a camera while someone pretending to be your friend grills you 60 Minutes–style for an hour, then forces you to sit there frame by frame watching the playback, pointing out every place you screwed up. If your ego survives that beating, congratulations, you're ready for primetime. But the result? A parade of plastic people on stage, delivering the most harmless nothing-statements imaginable.

Andreessen put it bluntly: the first rule of old media was "never be interesting" — being interesting was the most dangerous thing you could do.

After 2017, the Old Media Playbook Broke Completely

Andreessen recalled that from 1994 to 2017, he did a ton of traditional media, and 90% of the time he felt it was worthwhile — journalists gave him the chance to tell his story, and the public generally thought tech companies were cool, their products were fun, and the industry was good for America.

Then 2017 happened, and everything changed.

He believes traditional media originally had two missions: objective reporting and "speaking truth to power." The second mission gradually consumed the first, sliding from objectivity into activism — journalism with an agenda. Eventually, "speaking truth to power" effectively became "wielding power over truth" — using media's institutional clout to intimidate you into not saying what you actually think.

His advice to founders is blunt: Getting a story you're happy with through traditional media is now essentially impossible. You might pull it off once in a while, but you can't build a strategy around it. One success in 99 attempts is just the exception that proves the rule.

In the New Media Era, the Brand Is a Person

Andreessen offered a historical explanation. Before the 1930s, companies were named after their founders — Ford Motor, Edison Electric. Henry Ford probably never considered naming his company anything else. Then broadcast radio and television arrived, and media became radically centralized — three TV networks, a handful of major newspapers, all information squeezed through an impossibly narrow pipe.

Corporate branding emerged as a way to compress information into the smallest unit that could fit through that straw. International Business Machines, General Electric, General Motors, then Procter & Gamble's Tide, Colgate, Coca-Cola — abstract brand names replaced human names. Everyone assumed this was just how the world worked, but Andreessen argues it only held true in a world of centralized media.

Now centralized media is collapsing, and branding is shifting back to individuals. When people talk about SpaceX, they don't say "what's SpaceX up to" — they say "what's Elon up to." Palantir is Alex Karp. Anduril is Palmer Luckey. The companies winning the marketing war all have one person as the brand.

This means if you're the CEO, you probably can't sit this one out. It could be a co-founder, it could be another key figure, but it has to be someone who will permanently stay with the organization — not a VP of Marketing who's gone in three years.

Presidential Candidates Have to Go on Joe Rogan — Why Should Founders Be Exempt?

Andreessen used the 2024 U.S. election as an example: before that cycle, the answer to "Does a presidential candidate need to go on Joe Rogan?" was "Absolutely not." After 2024, the answer flipped to "100% yes." In the Democratic post-mortem, Kamala Harris skipping Rogan was identified as one of the critical mistakes. Every practitioner now believes the next candidate must do Rogan — and doing Rogan means sitting there for three hours, talking about anything.

That's the new bar. You have to be the person who can do that. If you're not, in Horowitz's words, "you've put a ceiling on your entire opportunity."

You Need People to Hate You, People to Love You, but You Can Never Be Boring

Horowitz shared a counterintuitive principle about brand building: You need people who hate you, people who love you, but you must never let anyone feel indifferent. Neutral and lukewarm equals boring, and boring means getting drowned out.

Once you reach a certain scale, attacks are inevitable. Horowitz said he used to tell colleagues who were anxious about negative coverage: nobody writes puff pieces about Rupert Murdoch or Elon Musk anymore — at that level, being hated is the norm, and it means you're doing something that matters. If your reaction is "I need to tone it down so everyone likes me," you've just killed your marketing.

A significant part of a16z's brand was built by punching back. Horowitz mentioned his proudest piece of writing — a response he published after a clash with the New York Times over the Instagram deal. It became his highest-traffic article ever. "Everyone loves a fight, and fights are great for brand building — but you have to pick the right opponent. Don't build someone else's brand for them."

Don't Start from Yourself — Start from the World

Andreessen described the most common and most fatal narrative trap he's seen.

The VC version: "I met this amazing founder, we took a walk, he called me a few times, and I finally convinced him to let me invest." The founder version: "We're a brand-new company, we're passionate, our product is great, please try it."

His verdict: these stories make you want to stab yourself in the neck. No specifics, no tension, fake humility — "I'm humbled by..." is the most dishonest form of false modesty on earth. You're not humbled at all. Worse, these stories are completely interchangeable — every startup, every VC can tell the exact same version.

You and your startup are not an interesting story in themselves, but there is almost certainly an interesting story connected to your company. That story is about something fascinating happening in the world, and your company just happens to be at the center of it.

Alex Karp's Masterclass in "Outward Thinking"

Andreessen called Palantir CEO Alex Karp the "grand wizard" of this strategy. If you watch Karp's interviews, he almost never talks about Palantir itself — the only two things he says about Palantir are "ontology" and "orchestration," two words nobody understands.

But it doesn't matter. Because Karp talks about the future of America's military, superintelligence, neurodiversity — he talks about the most interesting things happening in the world, and because he's Palantir's CEO, the association forms naturally. When something happens involving U.S. defense, AI, or geopolitics, Karp is the first call media makes, because he's the one who's been publicly engaging with these topics all along.

Flexport founder Ryan Petersen is another example. He didn't talk about freight — he talked about how global supply chains were collapsing during the pandemic, how people were facing a hunger crisis. The result? He became the guy on 60 Minutes in a helicopter, explaining to the nation "those ships that will never reach shore." He didn't even need to say "use Flexport" — the connection was implicit.

For companies doing enterprise sales, this is especially critical: Are you important enough to get a direct meeting with a client's CEO? To sit down with the Secretary of Defense? To walk into the White House? An obscure little company can't do any of that, but a founder who consistently speaks about major issues in public discourse can.

Get the Message Right Before You Worry About Distribution

Gabby identified the two most common mistakes founders make, and they're interconnected.

The first: rushing to distribute before the message is polished. Distribution is just a multiplier on your message — if the message itself is wrong, you're just amplifying a mistake. The worst-case scenario is you finally get on Joe Rogan and say the wrong thing.

The second: starting from inputs instead of outputs. Successful companies are especially prone to this, because there's so much interesting stuff to talk about — milestones, mission, product features — so they dump it all out and hope the audience picks out what matters. The right approach is the reverse: What kind of enterprise customer do you want to land? What kind of engineer do you want to recruit? What do they need to believe about you? Work backward from that endpoint to craft the right message.

Horowitz added a critical point: this directly determines how you build your team. If your team can't engage with you at the message level and is just passively executing "go market this," you're handicapped from day one. The team has to be involved in crafting the story, not just handling distribution.

For Your New Media Team, Hire Storytellers — Not Old Media Veterans

Horowitz's hiring advice was direct: if someone spent a decade in old media, the transition to new media is brutally difficult, because the two games have completely opposite rules. Every old media skill — vetting journalists, prepping Q&A, managing risk — works against you in new media. Very few people can make that switch.

When a16z hires, the core metric is: Have you done it yourself? What brand have you built, what audience have you assembled? Alex Danco, Henry, Brent — these people were previously product managers, founders, or investors, but they were obsessed with public discourse — they listened to podcasts, they wrote their own stuff. Even before they were formally doing this work, you could see the capability in their "proof of work."

a16z believes the most essential skill for a new media team is storytelling — not marketing-speak storytelling, but the ability to take something that actually happened and organize it into a narrative with a beginning, tension, and a resolution that makes you want to read it start to finish. The gap between great and terrible at this skill is enormous, and the people who can't do it will never learn.

Journalists See Only News and Propaganda — But the Real World Doesn't Work That Way

Horowitz pointed to a deep cognitive disconnect. In the eyes of traditional media professionals, only two things exist in the world: news and propaganda. Traditional media does news; everything a company produces is propaganda — "just marketing."

a16z flatly rejects that binary. On one hand, traditional media itself rarely delivers the objective reporting it claims to. On the other, companies' motivation for communicating directly isn't purely "selling things" — you genuinely want people to understand who you are, what you're building, and what's happening in the world around you. When they say "storytelling," they don't mean making things up — they mean "this is what's actually happening," just told in a way that's compelling and has narrative tension.

Andreessen mentioned that they had published a piece about SpaceX that day. It carried his name, but the credit belonged entirely to the team. He said the feedback was "this is the best SpaceX analysis ever written," because it simply laid out the facts as they are.

This Is a Learnable Skill

One final point: this isn't a talent — it's a skill. Andreessen said if you watch Alex Karp's early interviews, he's a completely different person from who he is today. Palmer Luckey, Elon Musk, and Ryan Petersen's early tweets and interviews are nowhere near their current level either.

He even used Donald Trump as an example — watch Trump's interviews from the '80s, and it's very standard old media style: restrained and proper. Then he figured out the new media playbook, became wildly entertaining and endlessly provocative, and that was the core engine behind his popularity.

Even at the highest level, this ability is developed, not innate.