Naomi Gleit on Clarity, Canonical Docs, and Disagreeable Givers

The Canonical Doc Problem

Naomi Gleit has worked on hundreds of projects at Meta over 19 years. When she ramps up mid-stream, the first thing she looks for is the one source of truth. What she often finds is chaos: five people, five different answers, hundreds of scattered documents with no clear hierarchy. This, she believes, is unacceptable.

I really believe in frameworks for things. That helps drive extreme clarity. I work on a lot of different projects. A lot of times I'm ramping up mid-project. I'm like, where can I learn what I need to learn about this project? I ask 5 different people, get 5 different answers. That is unacceptable.

— Naomi Gleit

Her solution is ruthlessly simple: every project needs one canonical document. Everyone on the team should know exactly where it lives. That document contains everything you need to know about the project and links to every supporting artifact. No treasure hunts. No conflicting narratives. Just clarity.

This obsession with frameworks and canonical sources isn't just bureaucratic tidiness. It's a forcing function for alignment. When you can't agree on where the truth lives, you can't agree on what the truth is. And when a company scales from 30 employees to 86,000, misalignment becomes existential. Gleit has seen that scale from the inside, longer than anyone at Meta except Mark Zuckerberg himself.

Cold Calling by Showing Up

Gleit didn't stumble into Facebook. She wrote her Stanford senior thesis on why Facebook would beat competitors like Friendster. She saw product-market fit in real time: Stanford students were obsessed, colleges were lining up on a waitlist, younger siblings were begging for access. So she walked into the office above Jing Jing's Chinese restaurant at 443 Emerson Street in Palo Alto and asked for a job. There wasn't one. She came back. And back. Five to ten times, by her count.

Eventually, she interviewed to be Sean Parker's personal assistant. She didn't get it. A few months later, a marketing role opened. She took it, even though she also had an offer from LinkedIn. The marketing job wasn't the dream—product management was. So she started volunteering on the second floor after hours, where the PMs and engineers sat, while her official desk was on the third floor with the business functions.

By the time that I actually applied formally to be a product manager, I had been doing the job voluntarily, almost informally for a few months. I picked up all the stuff on my desk, put it in a box, walked down to the second floor once I got the job to become a PM. And when I got to the second floor, I distinctly remember everyone on the second floor standing and clapping.

— Naomi Gleit

The lesson isn't about luck. It's about refusing to quit. If you're convinced a company is a rocket ship, don't wait for the perfect seat. Get on board, then create your own luck by doing the work before anyone asks you to.

Disagreeable Givers Run the Best Orgs

Mark Zuckerberg's small group—the core leadership team at Meta—doesn't map to the org chart. It's built around the people leading the most important projects, regardless of reporting structure. What makes it unusual is tenure. Many members, like Gleit, have been at Meta for close to two decades. They're not climbing a ladder anymore. They're motivated by mission, not promotion.

Gleit calls them "disagreeable givers," borrowing a framework from Adam Grant. In a two-by-two of agreeable/disagreeable and giver/taker, the most dangerous person is the agreeable taker: pleasant, well-liked, but fundamentally self-interested. The most valuable? The disagreeable giver. They'll challenge you, push back, create friction—but they're doing it because they care about the outcome, not their own advancement.

The most dangerous kind of person to have in an organization is an agreeable taker. And what that means is an agreeable person, super nice, everyone likes them, really easy to get along with, but they're a taker and maybe their motivation is more self-interested rather than what's best for a company.

— Naomi Gleit

Small group is full of disagreeable givers. That's why it works. The long tenure means trust is high, egos are managed, and the hard conversations happen faster. When everyone is already bought in to the mission, you can skip the posturing and get straight to the truth. This is the machinery that runs an 86,000-person company.

The Learn-It-All CEO

The world thinks Mark Zuckerberg transformed over the past few years—new hair, new vibe, new coolness factor. Gleit thinks the world is just catching up. She's known him since he was 19 or 20, running a dorm room project. The Mark people see now, with the long hair and the chains and the big shirts with his own phrases, is the Mark she's always known. The gap between perception and reality is finally closing.

What changed? Zuckerberg is a learn-it-all, not a know-it-all. He used to do annual challenges: learning Chinese, getting 8th-grade fluency in a year. Guitar. MMA. Public speaking. He was scripted and careful early on because he wasn't confident. He upskilled. Now he's comfortable, and people can see who he really is.

I've always said that there is the biggest gap of anybody I know between what people think of Mark and who Mark really is. And so, I think this is the Mark that I've known for the past 20 years and the world is finally getting to see what I've been lucky enough to see.

— Naomi Gleit

Gleit also knows a side of Zuckerberg most people don't: he's a thoughtful friend. In 2014, when she was going through a breakup, he asked her to co-teach a class with him in East Palo Alto. The CEO of Meta, teaching middle schoolers how to build a business. They wrote four life lessons on the chalkboard: love yourself, only then can you truly serve others, focus on what you can control, and for those things, never give up. They made stickers for the students. Gleit kept the lessons for herself.

Product Managers Are the Conductor

Gleit is Head of Product at Meta. A few hundred PMs report to her directly; a few thousand exist across the company. She doesn't manage them all, but she feels responsible for the entire PM community. That includes PM performance, culture, onboarding, and training. She sees PMs as an exponential lever—the highest point of leverage for getting things done and hitting company goals.

Her own path to PM wasn't standard. She wasn't technical. She had a Bachelor of Arts, not a computer science degree. But she wanted to build, and she knew PM was the right function. So she volunteered her way into the role, picking up program management tasks and giving product feedback until she'd effectively been doing the job for months before officially applying.

When she finally walked from the third floor to the second floor with her desk contents in a box, Boz and the rest of the team gave her a standing ovation. She'd earned it. The lesson: if you want to be a PM, start doing PM work. Don't wait for permission. Show up, ask for projects, build the muscle. By the time you ask for the title, you'll have already proven you can do the job.

Nineteen years later, she's still in it. Too busy to reflect, too focused on the to-do list. Her 20-year "Faceversary" is coming up. Maybe then she'll take a moment to look back. But probably not for long.