LLenny's Podcast
← All episodes
Noam Lovinsky (Grammarly, Facebook, YouTube, Thumbtack)17 March 2024

The happiness and pain of product management

8Frameworks
15Insights

Episode overview

Lenny Rachitsky speaks with Grammarly CPO Noam Lovinsky about lessons from building and turning around products at YouTube, Thumbtack, Facebook, and Grammarly. Lovinsky explains why product leaders should prioritize the broader business, diversify growth channels, and create distinct incentives and operating systems for zero-to-one teams. He also argues that career choices should maximize authentic energy, learning, and manageable discomfort.

Key ideas

  • Advocate for the organization’s highest priorities even when that means ending your own project or accepting another management layer.
  • Healthy marketplaces can prioritize customers first, suppliers second, and the company last because better customer demand ultimately benefits supply.
  • Reliance on one growth channel is fragile; Thumbtack’s SEO downturn demonstrated the need to develop paid, referral, and other organic channels.
  • Growth can conceal product and marketplace weaknesses, while a downturn forces teams to identify the real constraints and rebuild the core experience.
  • Zero-to-one incubators inside large companies need different time horizons, incentives, infrastructure, and permission to do work that does not initially scale.
  • Grammarly’s low-effort, workflow-embedded assistance and bootstrapped focus on revenue helped produce a durable consumer subscription business.
  • Career opportunities should stretch a person through learning and struggle while retaining one or two established strengths as a foundation.
  • People should spend more time on work that feels authentic, creates energy, and supports sustained excellence.

Transcript available · source text is retained privately and is not published

Frameworks in this episode

People & resources mentioned

Attributed to the moment in the episode. Timestamps are approximate.

People · 10

  • Bryan SchreierMentionsone of the board members, Bryan Schreier at Sequoia

    one of the board members, Bryan Schreier at Sequoia, said it was the prettiest smile graph that he had ever seen.

  • Noam LovinskyMentionsNoam is currently chief product officer at Grammarly. Previously, he was an early PM at YouTube, where he spent 5 years

    Today my guest is Noam Levinsky. Noam is currently chief product officer at Grammarly.

  • Hunter WalkMentionsthis uh VC Hunter walk he's a founder of Home proof Ventures

    Hunter Walk, who's amazing, was leading the viewer team

  • Shishir MehrotraMentionsshashir marocha is the co-founder and ceo of koda before starting koda shashir led the youtube product engineering and design teams

    Shishir Mehrotra, who's also very amazing, was leading the advertising team.

  • Salar KamangarMentionsa guy named Salar Kamangar, who's also awesome Google's sixth employee

    a guy named Salar Kamangar, who's also awesome Google's sixth employee and just learned a ton of from him

  • Marco ZappacostaMentionsAt Thumbtack Marco the CEO used to say that it feels like we're running up hill and chewing glass

    At Thumbtack Marco the CEO used to say that it feels like we're running up hill and chewing glass

  • Whitney SteeleMentionsone of them Whitney Steel is is running marketing at Descript now

    one of them Whitney Steel is is running marketing at Descript now

  • Mike SchroepferMentionsShrep was the the the org leader that was that was supporting NPE at the time

    Shrep was the the the org leader that was that was supporting NPE at the time

  • Tony FadellCoinedHe co-created some of the most innovative and beautiful and popular products in history. The iPod, the iPhone, the Nest thermostat.

    I recommend Build by Tony Fadell.

  • Lenny RachitskyMentionstoday we've got another very special compilation episode something I've been pulling on more and more with the podcast and the newsletter

    How would you guess, Lenny?

Resources · 36

  • Sequoia CapitalMentionscompany

    one of the board members, Bryan Schreier at Sequoia

  • FacebookMentionscompany · Meta

    When you were at Facebook, you built what is called the new product experimentation team

  • GoogleMentionscompany · Google

    There were many times where Google leadership reconsidered the acquisition and should we like sell YouTube

  • YouTubeMentionswebsite · Google

    At YouTube, when you joined, my understanding is YouTube was losing a lot of money.

  • WhimsicalRecommendssoftware · Whimsical

    Give them a try and see how fast and easy it is to build clarity with Whimsical.

  • VantaMentionssoftware · Christina Cacioppo

    This episode is brought to you by Vanta.

  • SubstackMentionswebsite · Substack Inc.

    I spend a lot of time with the Substack teams. Been a very active advisor there.

  • ThumbtackMentionscompany

    Thumbtack had kind of tried and and stopped paid channels you know other organic channels like you know referrals

  • HimsMentionscompany

    another one David Shine is running product at at Hims

  • DescriptMentionssoftware

    one of them Whitney Steel is is running marketing at Descript now

  • AirbnbMentionscompany · Brian Chesky, Joe Gebbia, and Nathan Blecharczyk

    I actually led that effort at Airbnb. It took 3 years of my life.

  • LinkedInMentionswebsite · LinkedIn

    Just go to linkedin.com/podlenny to claim your credit.

  • InstagramMentionssoftware · Meta

    instead of Facebook having to buy the next Instagram and WhatsApp and all the things

  • New Product ExperimentationMentionsother · Facebook

    you built what is called the new product experimentation team.

  • WhatsAppMentionssoftware · Meta

    instead of Facebook having to buy the next Instagram and WhatsApp and all the things

  • Burning ManMentionsevent

    I don't know how many people know about Burning Man and how it works

  • ThreadsMentionssoftware · Meta Platforms

    things like threads and ideas like threads were in that team all of the time

  • NikeMentionscompany · Phil Knight and Bill Bowerman

    one of the organizations that that we talked to that I felt like was doing this in one of the best ways was Nike

  • DuolingoMentionssoftware · Luis von Ahn and Severin Hacker

    There's like Duolingo, Grammarly, and I know you're doing B2B also, but there's so few.

  • Google DocsMentionssoftware · Google

    you would have you know, grammar and spell checking in Google Docs

  • Microsoft WordMentionssoftware · Microsoft

    or grammar and spell checking in, uh, you know, Word

  • ChatGPTMentionssoftware · OpenAI

    there's there's ChatGPT, there's, you know, Microsoft Copilot

  • Microsoft CopilotMentionssoftware · Microsoft

    there's there's ChatGPT, there's, you know, Microsoft Copilot

  • UkraineMentionsplace

    When you're a team that kind of like starts out of Ukraine

  • LithuaniaMentionsplace

    my mom was from Lithuania.

  • KyivMentionsplace

    my dad was born in Ukraine. He was he's from Kiev.

  • OdesaMentionsplace

    I was actually born in Ukraine. Oh, wow. I know Odessa.

  • BuildRecommendsbook · Tony Fadell and Noam Lovinsky's wife

    I recommend Build by Tony Fadell.

  • FargoRecommendstv show · Noah Hawley

    I just finished the the last season of Fargo. And every single season of that of that series I think is fantastic.

  • For All MankindRecommendstv show

    I really I really like For All Mankind.

  • Apple TV+Mentionsproduct · Apple

    You've seen that on Apple Apple TV.

  • ArcUsessoftware · The Browser Company

    I was a very early user of Arc and I really love Arc.

  • GrammarlyUsescompany · Grammarly

    I I'm an active Grammarly user.

  • SpotifyMentionssoftware · Spotify

    you can subscribe to the show on Apple Podcasts, Spotify, or your favorite podcast app.

  • Lenny's PodcastCoinedpodcast · Lenny Rachitsky

    You can find all past episodes or learn more about the show at lennyspodcast.com.

  • Apple PodcastsMentionssoftware · Apple Inc.

    you can subscribe to the show on Apple Podcasts, Spotify, or your favorite podcast app.

Spot an error or want something removed? Request a correction or removal.

Insights & moments

The myth-busts, hot takes, explainers, and tools worth keeping.

Hot Take· 1

Hot Take24:00

Growth Masks All Problems

When a business has incredible growth, you can't truly tell what's working and what isn't — Thumbtack was quietly dropping demand on the floor due to insufficient supply-side liquidity, and no one treated it as urgent while triple-digit growth continued. When growth slows or goes negative, the whole tenor changes and real issues surface. Noam sees going through this as healthy, and now habitually asks what he'd prioritize if growth went negative.

  • Incredible growth hides which parts of the product are actually broken
  • Thumbtack was burning demand it couldn't serve, but growth masked it
  • Only when growth slowed did the org start diagnosing the true problems
  • PMs who've never felt the downside lose something in their career
  • Noam now asks 'what would I do if growth went negative?' as a prioritization lens

you know growth masks all problems like you don't really have an I think true understanding of like what is working well and what is…

Noam Lovinsky · 24:30

I often look at things and ask myself like okay what would I do right now if it went negative

Noam Lovinsky · 26:00
#growth#prioritization#marketplaces#product-management

Explainer· 2

Explainer39:30

'Find the Next Instagram' Is the Wrong Bar for an Innovation Lab

Judging Facebook's New Product Experimentation team by whether it built the next Instagram is like judging a group on whether it won the lottery. The team's real value was rebuilding muscles a giant org had lost — doing things that don't scale, running experiments with tiny user counts, and letting PMs and designers sit directly with customers. Even 'champagne' outcomes are rare; the 'beer/nice-dinner' outcomes and the mirror it held up to the org were themselves valuable, and it was a great recruiting tool.

  • Discovering an Instagram is closer to winning the lottery than a repeatable target
  • At Facebook/Google scale, running an experiment with 100 people is effectively impossible
  • NPE recreated the muscle to start small and do things that don't scale
  • It let PMs, engineers, and designers talk directly to customers — normally near-impossible at scale
  • It exposed what was broken about zero-to-one inside a giant org, and was a strong recruiting tool

I think in terms of was it a success or not because it didn't build the next Instagram I think is a little bit of…

Noam Lovinsky · 39:30

it is really hard for product managers engineers and designers to talk directly with customers it is like basically impossible

Noam Lovinsky · 42:00
#innovation#zero-to-one#facebook#product-management
Explainer52:00

Why People Love Grammarly: An AI Achievement Disguised as UX

Noam attributes Grammarly's success to how and where it works: you install it and it just makes you better, with no configuration, across every app and text box you use. It takes tiny effort to get real value and it's everywhere — you write in tens or hundreds of text boxes a day. Delivering that quality at that scale is really a huge AI achievement masquerading as a little UX innovation.

  • Install it and it makes you better — no configuration or workflow change needed
  • Assistance is pushed in the right moment across all apps and tabs; ignore it freely
  • You write in tens to hundreds of text boxes a day, so it's everywhere and low-effort
  • It's fundamentally 'a huge AI achievement masquerading as a little UX innovation'

Grammarly is one of the few products where you just install it and it makes you better.

Noam Lovinsky · 52:30

essentially it's like a, you know, uh, a an huge AI achievement masquerading as like a you little UX innovation, right?

Noam Lovinsky · 53:30
#grammarly#product-design#ai#ux

Story· 3

Story09:30

He Talked YouTube Into Killing His Own Project

Three months into YouTube (via acquisition), Noam told leadership they shouldn't put 50 engineers on the product he'd built, because the priorities warranted the team elsewhere — effectively negotiating himself out of a job. Instead of being punished, leadership wound it down and put him in charge of the creator focus area. The lesson: in a healthy org, advocating for what's best for the team and business pays off, and if it doesn't, that's worth learning early.

  • Noam told leadership not to staff 50 engineers on his own project
  • He did it at month three, effectively negotiating himself out of a job
  • Leadership rewarded the honesty by giving him a bigger role (creator focus area)
  • In a healthy team, advocating for the org's good is rewarded — if not, learn that early

I think that this team would would likely be better served elsewhere even though that was likely negotiating my way out of a job and…

Noam Lovinsky · 10:30

advocate for like what's best for the team. Advocate for what's best for the organization even if that means that it puts you at a,…

Noam Lovinsky · 12:00
#youtube#prioritization#career#leadership
Story12:30

Why He Asked to Be Demoted at YouTube

Struggling reporting directly to the CEO, Noam asked to report to peer Hunter Walk instead — atypically requesting to be layered under someone. He judged it would let him do better work, get better support, and be happier and more productive. He was right: Hunter became a great manager and Noam grew a lot.

  • Noam struggled reporting to CEO Salar Kamangar, whose questions felt 'from a different planet'
  • He proactively asked to report to peer Hunter Walk instead
  • Almost no one volunteers to be layered under someone — he framed it as how he'd do better work
  • The move paid off: better management, support, and personal growth

No one has ever come to me in my career and said, I would like you to layer me

Noam Lovinsky · 13:30

this is how I will do better work. This is how I will get better support.

Noam Lovinsky · 13:30
#career#management#youtube#self-awareness
Story28:30

How Thumbtack Rebuilt Its High-Friction Marketplace

Thumbtack's original model made customers answer questions, wait 24 hours, and then get quotes from pros who paid just to show up — high friction on both sides, and sometimes unfair to pros paying without real intent. The turnaround rebuilt the whole loop so pros could give instant quotes, changing the monetization model — the same request-to-book to instant-booking shift Airbnb made. Doing it across thousands of sub-marketplaces was what really turned the growth engine around.

  • Original flow: answer questions, wait 24 hours, then pros pay to quote
  • Customers just want to see who they can hire; making supply pay to appear added friction
  • Pros paying without high-intent leads turned into a bad loop
  • The fix mirrored Airbnb's request-to-book → instant-booking shift, across thousands of sub-marketplaces

What customers want to see is the supply. Like, tell me tell me who I can hire.

Noam Lovinsky · 30:00

the move from request to book to like instant booking.

Noam Lovinsky · 31:00
#marketplaces#product-strategy#monetization#thumbtack

Tool· 1

Tool1:04:00

Book Recommendation: Build by Tony Fadell

Asked for his most-recommended books, Noam gives just one: Build by Tony Fadell. Beyond being a good book, he recommends it partly because his wife co-wrote it with Tony. What inspired him was seeing how many times Fadell met failure before making the discoveries that now drive so much of what we use — a reminder to keep at the thing that gives you energy.

  • Noam's single top book recommendation is Build by Tony Fadell
  • His wife co-wrote the book with Tony Fadell
  • The book details how many failures Fadell hit before his big discoveries
  • Takeaway: keep at the work that energizes you — the discovery can come eventually

I recommend Build by Tony Fadell.

Noam Lovinsky · 1:04:00
#books#recommendation#perseverance#product

Takeaway· 8

Takeaway04:30

You Don't Need to Post Online to Win — Just Do the Work

Noam has been hugely successful in tech while barely tweeting, writing, or podcasting. He argues you should lean on what feels authentic and energizing rather than forcing an online presence because you think it leads to success. If you do what genuinely fills you up, you're likely to become one of the best people at it.

  • Noam gets deep on one thing at a time and finds broad multitasking (like public posting) hard
  • The best people often aren't tweeting or on LinkedIn — they're doing the work
  • Do what you like; you'll be far better at things that excite you
  • Leaning on what's authentic to you tends to make you one of the best at it

I get really focused on the thing that I am working on and get really deep in the thing that I am working on and…

Noam Lovinsky · 05:30

do what you like. You're you're generally going to be a lot better at the things that like really fill you up, that really get…

Noam Lovinsky · 07:00
#career#authenticity#personal-brand#product-management
Takeaway16:30

Prioritize What's Best for the Business, Not Just the User

A PM's job is to work backward from what's best for the business broadly — not always the user, and not the team's own goals. At Thumbtack, the ordering was customers first, pros second, Thumbtack last, guided by the insight that what pros most need is more (and higher-quality) customers. Guiding principles help navigate thorny tradeoffs like raising prices that hurt marketplace liquidity.

  • Best-for-user and best-for-business overlap but not always
  • Thumbtack's principle: customers first, pros second, Thumbtack last
  • What pros need most from the marketplace is more high-quality customers
  • Guiding principles help resolve conflicts and avoid local maxima (e.g. price hikes that kill liquidity)

So, it was customers first, pros second and then Thumbtack last.

Noam Lovinsky · 17:30

we we always just felt that what the pros need from us is more customers.

Noam Lovinsky · 18:30
#prioritization#marketplaces#product-strategy#thumbtack
Takeaway19:30

The Real Question Isn't When to Quit — It's When You Run Out of Stamina

Noam says what kills most projects and early companies is stamina, not money. High-ambiguity work with weak signal demands ungodly faith and resilience. The right time to move on is usually when a team runs out of steam and stops bringing its best self — often before it runs out of money.

  • Stamina, not money, kills most projects and early companies
  • High-ambiguity work requires 'faith and stamina' because signal is weak
  • A motivated team's inertia and quality are a completely different game
  • Sometimes a burned-out team just needs a change of scene or pace

I think the reality is is that what kills most projects most early companies is is stamina.

Noam Lovinsky · 20:00

Marco the CEO used to say that it feels like we're running up hill and chewing glass

Noam Lovinsky · 20:00
#resilience#startups#when-to-quit#product-management
Takeaway22:00

A One-Channel Growth Company Is Always a No-No

Thumbtack grew brilliantly on SEO — but SEO is live-by-the-sword, die-by-the-sword, and when Google changed its algorithm, growth went negative. Depending on a single growth channel is dangerous because it can dry up without warning. The fix is to diversify proactively — turn on paid, referrals, and other organic channels before the main channel breaks.

  • SEO is 'live by the sword, die by the sword' — Google changes hurt Thumbtack badly
  • A one-channel growth company is always a no-no
  • Diversify into paid, referrals, and other organic channels before trouble hits
  • Now Noam's 'paranoia' perks up whenever a product leans on a single channel

SEO is a sort of like a live by the sword die by the sword you know channel of of growth and I think that…

Noam Lovinsky · 22:30
#growth#seo#distribution#thumbtack
Takeaway32:00

The Disruptor Test: What Would Someone Starting Today Do?

Lenny shares the lens he used at Airbnb: ask what a fresh disruptor starting now would build — and for Airbnb it was obviously 'make it instant.' Noam adds that any product involving 'bits and atoms' is exponentially harder than pure-software products, and something as seemingly simple as 'make it instant' turns out to be incredibly deep and hard to pull off on an existing business.

  • Ask: if a disruptor started today to become the new us, what would they do?
  • For Airbnb the answer was obviously to make booking instant
  • Products involving 'bits and atoms' are exponentially harder than software-only
  • Simple-sounding changes like 'make it instant' are deceptively deep, especially mid-growth

you should be asking yourself, if somebody was to come into our space and disrupt us and start now to become the new Airbnb, what…

Lenny Rachitsky · 32:00

any product you work on that involves like bits and atoms is exponentially harder than products that just involve bits.

Noam Lovinsky · 32:30
#product-strategy#disruption#airbnb#marketplaces
Takeaway35:30

On a Leadership Team, No One Is a Bystander on Product Strategy

When growth goes negative, the CEO's staff — CFO, head of sales, head of people, engineering — must tackle product strategy together, even though they rarely do project work as a group. Just because 'product' is in your title doesn't make you the only one thinking about product strategy. PMs are the connective tissue, but the whole leadership team should feel their fingerprints are on the company's strategy.

  • CEO-staff peers rarely have natural ways to do project work together
  • In hard times, that group's ability to tackle problems jointly is critical
  • Having 'product' in your title doesn't make you the sole owner of product strategy
  • Everyone must feel their fingerprint is on the strategy, or it becomes 'their world vs ours'

no one had no one can be a bystander on product strategy.

Noam Lovinsky · 36:00
#leadership#product-strategy#org-design#collaboration
Takeaway46:00

The Incentive System Makes or Breaks an Internal Startup

The single biggest thing when building a startup within a big company is the incentive system — smart people game toward whatever it rewards. If you evaluate your zero-to-one incubator on a twice-a-year performance review, you've already killed it: it's the wrong time frame and creates adverse selection. Give teams startup-like autonomy over infrastructure, leveling, and pay, and think about real upside and longer time horizons — Nike's incubation lab was a strong example.

  • Smart people game toward the incentive system even unintentionally
  • A twice-a-year performance review is the wrong time frame for an incubator and kills it
  • Wrong incentives cause adverse selection in who you attract
  • Let internal-startup teams choose their own infrastructure and throw away code freely
  • Compete with people starting their own thing: think upside and startup-length time horizons (Nike's lab as a model)

if you're large organization and you do some performance management process like twice a year and that's how you're going to evaluate and incentivize people…

Noam Lovinsky · 46:30
#innovation#incentives#zero-to-one#org-design
Takeaway59:30

Take the Jobs That Stretch You — the Pain Precedes the Promised Land

You can't predict with certainty which job or bet will hit, but you can develop a nose for people you'll succeed with, and you can reliably find situations that stretch you. Noam has always prioritized roles that cause a lot of growth and learning — which is often painful — because on the other side of that pain is the promised land.

  • You can't forecast success with certainty, but you can develop a nose for the right people
  • You can reliably find situations that will stretch you
  • Prioritize roles that force growth and learning, even though it's painful
  • On the other side of that pain is the promised land

I always try to prioritize putting myself in positions that are going to cause a lot of growth and learning.

Noam Lovinsky · 1:00:00

on the other side of that pain, I think is is is the promised land.

Noam Lovinsky · 1:00:00
#career#growth#advice#learning