Matt Mullenweg, founder of WordPress and Automattic, joins Sam and Shaan to talk about starting WordPress at 19, turning down a $200M acquisition offer at 24, the open source philosophy behind his decision-making, the WooCommerce acquisition story, and the WP Engine controversy. He also shares unconventional practices at Automattic — like hiring via paid auditions and requiring everyone including the CEO to rotate through customer support.
Speakers: Sam Parr (host), Shaan Puri (host), Matt Mullenweg (guest, founder of WordPress / Automattic)
Introduction [00:00:00]
Sam: All right, today’s episode is special. We’ve got Matt Mullenweg. Matt founded a company called WordPress, which is used by something like 45% of all websites on the internet. We talked to Matt about a bunch of interesting things. Shaan, what did we cover?
Shaan: He had an offer to sell his company for $200 million when he was 24 years old. He turned it down. We asked him what that was like. We talked about some of the recent drama they’ve had. We talked about how they’ve been acquiring companies — they bought a small company in South Africa that turned out to be a billion-dollar-plus win. And he’s just a student of the game. This guy started the company when he was 19 and is still doing it. It’s become an absolute juggernaut.
Turning Down $200M at 24 [00:02:00]
Sam: Tell me if this is right, because it sounded almost too good to be true. In 2008 you had an acquisition offer — I think you were only 24 years old — for $200 million. At that point you’d only raised a million dollars, at a $3 million valuation. You were 24, you were going to be worth nine figures. You turned it down. But you also mentioned you didn’t have full control of the company at the time. What’s the conversation like with yourself when you’re considering turning down something that might make you worth over $100 million at 24?
Matt: You talked about your first million — technically on paper my first million was that first funding round, where I owned maybe half the company and it was worth $4 million. But that’s paper money. I was still eating ramen and Mountain Dew pizza, living a very broke San Francisco college-kid life.
Matt: In 2008, about two and a half years in, someone tried to buy us for $200 million. The investors did something which is now common but was forward-looking at the time — a secondary. They said, “Wow, we’re 20 people, two and a half years in. A $200 million exit would be pretty amazing. But we think this could be way bigger. Let’s build.” So we turned the acquisition into a valuation and raised a new round. I sold some stock myself. That was my first liquid million — around 2008, when I was 24. It was a step change. I paid off my credit cards, bought my mom a house. All that stuff you dream about. And it removed some of those early economic pressures so I could really focus on the business and swing for the fences.
Sam: I hear these stories — Zuck turning down a billion from Yahoo — and I think you’re a better man than me. I don’t think I could have resisted. Was that an easy decision?
Matt: Not easy at all. You have to seriously consider these things. As a fiduciary, you have a responsibility to shareholders to consider every acquisition. We’ve had other acquisition offers as recently as this year. I think you have to ask yourself: will the mission we’re doing be accelerated by this transaction, or will it be hampered?
Shaan: Reddit was one you looked at too, right?
Matt: I really wanted to buy Reddit. I couldn’t convince my board — they thought it was too outside our focus. But yeah, when they had four employees and were for sale, sitting in the Wired offices on Third Street in San Francisco, I thought it was really cool.
How WordPress Grew [00:10:00]
Sam: When you created WordPress it seemed like it took off within a year. I was using it around 2010 or 2011 — a Tennessee college kid who’d already heard of it.
Matt: First version was 2003.
Sam: Oh wow.
Matt: I wrote a blog post about this — what people see as overnight success is often a thousand days of irrelevance. At one point there was a joke that WordPress had more developers than users. The first few blogs were just ones I set up for friends in high school. We used to do upgrade parties — a new version would come out and I’d open up my apartment, go to Costco, buy some booze and pizza, and say, “Come over and I’ll upgrade your site for you.” Really just doing the work.
Matt: It looks like overnight success later, but we had some genuine breakout points — like when Movable Type changed their license. Fortune favored the prepared. We’d put in a lot of grind, community building, contributions, code — in many many days before that moment.
WordPress as the Internet’s Dark Matter [00:15:00]
Sam: What’s crazy is — a few years ago I read that WordPress was used by like 20-30% of all websites. Now it’s 40%. And I look at this and think: are you guys the most under-monetized business on earth? I used WordPress and WooCommerce at my old company. The WooCommerce license was something like $300 a year. The product was making many millions of dollars. I’ve got a friend who made a hundred million dollars off a $300-a-year license. You have to be the least monetized company there is.
Matt: The way I put it is WordPress is almost like the dark matter of the web. The ecosystem of WordPress is probably at least $10 billion a year in revenue now. My company Automattic is about 5% of that. But if you add up all the companies and all the people — not even counting things like goods and services sold through WooCommerce, which was over $30 billion last year — it’s massive. And more than half of Automattic’s revenue now comes from things that aren’t just WordPress. We have some really cool mobile apps like Day One or Pocket Casts, and a new one called Beeper.
Sam: What are the top two acquisitions, like Berkshire-style where a couple key moves made the business?
Matt: Our most successful is probably WooCommerce. It came a lot from studying platforms. I did a deep read on Microsoft — for every dollar Microsoft made from Windows, there was $20 made by the Windows ecosystem. That ratio is similar to ours. True platforms create far more value than the host platform.
Matt: WooCommerce started as a small company — about 40 people based out of South Africa. It started as a theme company called WooThemes, and they developed this fork of another open source e-commerce plugin. They built it just to sell more themes. The code was honestly pretty crappy when we first looked at buying them, but they’d done such a good job building something people wanted. Product-market fit was excellent, even if the code wasn’t scalable.
Matt: When we did move to acquire, there was private equity trying to buy the same plugin. They chose us because of our culture and mission. We were able to take what we’re great at — engineering and scalability — and apply it to what they’d built brilliantly. That became the thing that did $30 billion-plus in goods and services last year.
Open Source vs. Shopify [00:25:00]
Shaan: The backhanded compliment: 43% of the internet runs on WordPress-related products. That’s an incredible number. And on the other hand, Shopify’s market cap alone is $150 billion. A ruthless capitalist would say, “Matt, you’re doing all this work, and the closed-source Shopify variant of e-commerce is worth $150 billion while Automattic is worth several billion.” What do you take away from that?
Matt: There are definitely things that are easier in a proprietary closed ecosystem. Shopify is really great at forcing people to use their payments, for example. In WooCommerce you can use ours but you can also use many other payment providers. Their average revenue per subscriber is probably about 10x ours.
Matt: But how I think about it is short-term versus long-term. I have this philosophy of open source that I believe in morally. I want all the work I do to increase the amount of freedom and liberty in the world. That’s why I’ve dedicated my life to open source. Open source software has a Bill of Rights attached to it — the freedom to use it for any purpose, to see how it works, to modify it, to redistribute modifications. The Four Freedoms of the GPL. That’s a moral decision.
Matt: If Shopify changed their policies tomorrow, their customers are stuck. No recourse. With open source, I could become evil or Automattic could become a terrible company — you would still own all the code. WordPress and WooCommerce belong just as much to you as they do to me. That freedom and liberty is better in the long run.
Matt: Open source has a slow burn. Often it’s slower to start, but over time it builds this compounding momentum that becomes unstoppable. It’s already 10x the number two in the market. And it forces proprietary companies to be more open. Android exists and Apple is probably more open than they’d be otherwise. So even if we don’t make as much money as Shopify now, check in 10 or 20 years.
The WP Engine Controversy [00:35:00]
Sam: So Matt, you’re clearly this thoughtful, almost soulful entrepreneur. You’ve been building this thing since you were 19. And then you had this villain arc that people tried to paint on you over the last year with the WP Engine drama. I couldn’t believe it. If I was going to bet on who was the least drama-attracting founder, it might have been you.
Shaan: I didn’t follow it too closely. But quick reaction, Sam?
Sam: I’m a WordPress user. And I’m friends with Jason Cohen of WP Engine. You guys had a fight. I was actually shocked, Matt — you felt insulted and you insulted back, which surprised me. You don’t seem like someone who would ever insult someone. And I guess your perspective is they were coming after everything you made. But I was surprised you were going as hard as you were.
Matt: A failure mode — and this can kill many open source projects — is when they get taken advantage of. Like a schoolyard bully, you kind of have to stand up for yourself. Actually, if you look at the history of WordPress, there have been maybe four or five times where I had this kind of villain arc — we had a fight to protect our principles and the sustainability and future of WordPress.
Sam: Give the one-minute summary of what happened.
Matt: It’s an ongoing legal battle so I can only say so much. Basically, there’s a company called WP Engine — I think Jason Cohen is awesome, by the way. But in 2019 they were bought by a private equity firm called Silver Lake, and over the subsequent five years they became what I would say is more parasitic of WordPress. They also created a lot of confusion in the marketplace with how they were marketing themselves, in a way that was threatening our trademark. People would say “WordPress Engine” and they wouldn’t correct them — people thought it was our official company. I frequently got support requests for WP Engine. Even close friends who were WP Engine customers thought it was my company.
Matt: For about two years prior to the fight we were trying to partner with them and resolve the trademark issues. They just weren’t responding.
Shaan: On the contribution piece — does somebody have to contribute to the open source project? Is that a rule or a suggestion? Like, you’re at church and should put something in the tray, but technically you don’t have to?
Matt: We have a program called Five for the Future — it’s completely voluntary. If you’re building a business on WordPress, we say if you can allocate somewhere between 1% and 5% of your profit or revenue or time, and put that back into Core WordPress — the open source project accessible to everyone — that’s what’s made us sustainable. What’s made us thrive more than other great CMS projects like Joomla or Drupal. I think this is also great self-interest.
Matt: WP Engine is somewhat unique in the ecosystem because pretty much every other company contributes. In fact, old versions of WP Engine’s website were very supportive of this — they said they’d dedicate two to four full-time people to core. Fast forward to 2024, they had less than that.
Sam: I didn’t follow the story in depth, and I don’t need to. Common sense to me — whose side am I on? The private equity-backed company that sounds like it was made by the WordPress people but isn’t? Or the founder who’s been working on this for 20-plus years, open-sourced it, it’s used by everyone, captures a tiny bit of the value along the way? Pretty obvious.
Matt: I expect this to resolve in the next few months. These controversies come and go. If you go to my Wikipedia page right now, their PR firm has a whole paragraph about this. In five years maybe it’ll be a sentence or not even on there at all.
How Automattic Hires [00:50:00]
Shaan: On your blog — or maybe on the Tim Ferriss podcast — you wrote about how Automattic has roughly 2,000 people, and you tried a bunch of different hiring methods. And you said two interesting things. One: the people who are the best writers are often the best hires — not PhDs, not master’s degrees. There’s a correlation between writing ability and the quality of hires. And second: you used to hire people just through chat — text-based, asynchronous. Do you still do that?
Matt: For some roles we might do a Zoom, like sales, where seeing how someone interacts is important. But for a lot of roles, written communication is going to be the primary thing anyway. So a lot of our hiring can be completely asynchronous and text-based. For the first thousand or so hires, I did a final chat for every single person.
Sam: Slack and DMs, basically?
Matt: Yeah. Slack when Slack was invented. Before that, Skype, AIM, IRC.
Shaan: You said “auditions, not interviews.” What does that mean?
Matt: We do a trial project. We actually hire people on a standard $25-an-hour contract and pay them to do some work together. We have screens — resumes, a brief interview — but then we say, “Let’s do some work.” There are various versions of this for different roles. We’ve done sandbox versions; we’ve also done it where they were actually talking to real customers — a support person answering real tickets.
Matt: We’re smaller than the big tech companies but compete with them for talent. So we need to find people who might be overlooked by a Meta or Google and give them an opportunity to have just as much impact. We also pay people the same salaries regardless of location. Same work, same pay — globally. If you’re doing the same job in Pakistan as in California, you get paid the same. I think that’s the moral position, and I think it’s the future of work.
Customer Support Rotation [01:00:00]
Shaan: You also do something interesting where everyone in the company — including yourself — works customer support one or two weeks a year?
Matt: Your first two weeks at Automattic, regardless of role — new CFO, chief legal officer, whatever — you’re doing customer support. Then once a week a year you rotate back in.
Matt: Why? The closer a business is to its customers, the more successful it generally is. It’s very easy when you’re running something on the internet for customers to become numbers or stats on a Looker dashboard. Getting back to real individual customers — each one of those 5,000 daily signups is a person with a story. You learn a lot about your product. Eric Ries and Steve Blank talk about getting out of the office and meeting customers. I’ve heard that leaders at Salesforce spend a quarter to a third of their time with customers even at that scale.
Sam: Any epiphanies from doing this yourself?
Matt: Just a few days ago I spent 30 to 45 minutes with the gentleman who reviews expenses at the company. I’d gotten feedback that some questions he was asking people were coming across as aggressive. I wanted to understand the tools he was using, see the work firsthand, and talk about the culture. Not just “you didn’t have a receipt” but explaining the principles: “If we don’t have this receipt, our internal audit firm might question it, and that creates this downstream issue.” Bedside manner, basically.
The Birthday Alarm Story [01:06:00]
Sam: When I first moved to Silicon Valley, I came to work with this guy Michael Birch. He represented everything I wanted — he’d already built successful tech companies. I was a 23-year-old who wanted to make it. Ready to learn the dark arts, the growth hacks, all the high-level strategic thinking. And the first week he puts me on Birthday Alarm — some company he’d started back in 2001.
Shaan: Birthday Alarm is still going?
Sam: 25-year-old company now. So I get curious and ask him the story. He said: “My very first startup — I quit my job, my wife was pregnant, I wanted to build an internet company. I tried to build a self-updating address book. You know your friend moved? Wouldn’t it be cool if they updated their info in one place and it updated everywhere?” Shawn Parker tried the same thing. He spent nine months heads-down coding this. It wasn’t going anywhere. But because he was a one-man show doing customer support too, he noticed that over and over again people kept thanking him for the birthday reminder feature — this throwaway thing he’d built in. This was before Facebook. He threw away the whole product, renamed it BirthdayAlarm.com. Expected it to go nowhere. That was the thing that took off. At that time it had generated about $20 million of pure profit for him and his wife. Still generating a few million a year.
Matt: And only came because he was answering support tickets and got curious.
DeepSeek and AI [01:14:00]
Sam: Last question — your quick take on DeepSeek.
Matt: DeepSeek is a really cool model. Every model has a vibe and a way it’s tuned, and the DeepSeek vibes are very cool. What I’m most excited about as an open source guy is that they actually open-sourced the model and wrote amazing papers about how they built it. Open weights.
Matt: At my company I’d say don’t use deepseek.com — it’s hosted in China — but you can run the model yourself locally, or access it through Perplexity, which is hosted in the US. They made some real advancements in how they train, how they run things, how they did memory interconnects, working within constraints. They shared it all. That’s the key thing — OpenAI had probably figured some of this out too but hadn’t shared it publicly.
Sam: Do you think they actually only spent $5 or $10 million to train it, like they claimed?
Matt: That might be true for the final training run specifically, but obviously they invested a lot in other things leading up to it. I’m sure there’s some PR spin in there. But the things they made — some amazing advancements — you can verify them. And they open-sourced it all under a true open source license, not like the Llama license. Also check out Qwen from Alibaba — people are sleeping on that one. Really great model.
Sam: Matt, thanks for coming on. Thanks for sharing everything about WordPress. It’s been a pleasure.
Matt: Yeah, appreciate you guys.
Sam: That’s the pod.