By playing, you agree to uListen's Terms and Privacy Policy and YouTube's Terms of Service.
a16zGrowth vs Efficiency: Can You Have Both?
Summaries, chapters, transcripts, and FAQs on this page are made by uListen, not YouTube.
CHAPTERS
- 0:00 – 2:59
Why Growth Is Harder in 2023: Framing the Episode
Steph Smith sets out the paradox that building a product is easier than ever while growing one is harder, as old channels erode. She introduces the panel: Gina Gotthilf, Kieran Flanagan and Bryan Kim.
- •Paid platforms, SEO, social graphs, product-led growth and virality have all been disrupted
- •The episode asks how to balance efficiency and profitability while still growing
- •Gina Gotthilf led growth at Duolingo from 3 to 300 million users with no paid budget
- •Kieran Flanagan was SVP of Marketing at HubSpot and Bryan Kim is an a16z partner who held leadership roles at Snap
- 2:59 – 6:56
What Growth Still Works: Measurability, Channels and Repetition
Kieran Flanagan describes a shift back toward indirect causation as short-form video, community, referral and word of mouth gain importance. Gina Gotthilf argues that the core framework of understanding what makes people tick has not changed, even as channels rise and saturate.
- •AI may be a tailwind for 12–18 months, and search and paid ads still hold value despite lower click-through
- •Marketing became measurable with the internet, but founders still want $3 back for every $1
- •Over-indexing on measurement pushes everyone to the middle and drains creativity
- •Gina says to focus on the moments people care about and on reaching one person or channel that reaches many
- •Channels like New York subway ads and notifications rise and saturate, so finding cheaper new ways to impact people matters
- 6:56 – 11:39
Growth as a Retention Equation: Product vs. Distribution
Bryan Kim argues growth is cumulative, depending on how many people keep coming back, which ties it to product value. He then separates product from distribution and surveys which distribution channels still work, including Reddit, Discord and AI-driven ad creative.
- •Growth is like an equation: returning users plus new users minus churn plus resurrected users
- •The biggest drop-off is typically day zero or day one, if users don't reach the aha moment
- •Product-led growth comes down to core product value delivered repeatedly
- •Subway, TV and radio ads can still work depending on the audience, and Reddit and Discord suit gamers and niche communities
- •Meta's AI could automatically vary ad creative and restore paid-ad efficacy after ATT
- 11:39 – 16:17
Moats in an Era of Easy Copying: Duolingo and Zapier
Steph asks whether a first-mover advantage matters when products are easy to copy. Gina explains that Duolingo's product is its moat because of its accumulated data and learning, while Kieran discusses Zapier's moats and how AI may shape B2B differentiation.
- •Metrics now make it clear whether a product is good, and products must constantly learn and evolve
- •A moat can be a brand, such as Kim Kardashian launching a shoe, or the product itself, as at Duolingo
- •Competitors copied Duolingo's fonts and colors but lack its data, history and team
- •Kieran says first-mover advantage helps only with a healthy retention curve, and AI is an unlock for product-led growth
- •AI may skew toward incumbents with distribution, and distribution, retention, network effects and community remain moats
- 16:17 – 17:22
Trust as a Moat (and How to Say Zapier)
Gina Gotthilf says community is Latitud's moat and that trust is a moat that takes time to build, which applies to B2B too. The conversation then turns to a lighthearted aside on how to pronounce Zapier.
- •Gina says trust underpins Latitud's community moat
- •A trusted brand leads people to trust a company's new features and solutions
- •Competitors can offer the same product but can't easily replicate built trust
- •The group jokes about the pronunciation of Zapier
- 17:22 – 21:58
Consumer Distribution Is Ossified: Breaking Through Incumbents
Bryan Kim explains why consumer growth is harder: people download fewer new apps, App Store discovery is difficult, and incumbents hold network effects and data. He says the only way to stay ahead is to keep getting new products to hit the zeitgeist.
- •People are downloading fewer apps and App Store discovery is harder, so platforms have ossified
- •Incumbents have network effects, logins and data, and they copy everything down to the font
- •The three boss levels are getting through the noise, building a retentive product and repeatedly hitting the zeitgeist
- •First hit is hard, second is rare, and a third makes a generational company
- •Staying ahead depends on founder culture and not resting on laurels
- 21:58 – 24:59
Snap's Repeated Innovation: Disappearing Images, Stories and Lenses
Asked for an example, Bryan Kim walks through how Snap kept producing hits that were then widely copied. He stresses the long-term, multi-bet system of innovation behind it, and also mentions BeReal.
- •BeReal is cited as a portfolio company that has done this with certain features
- •Disappearing images turned the medium into the message
- •Stories became a second zeitgeist hit and are now copied everywhere, including LinkedIn
- •Lenses addressed people's dislike of the front-facing camera and also became ubiquitous
- •Success needed several simultaneous bets, thoughtful choices about what to kill, and a founder mentality
- 24:59 – 28:13
Fostering a Culture of Speed and Learning in Austerity
Steph asks how to foster a shipping culture when budgets are tighter. Kieran says many founders are glad to return to fundamentals and shipping, while Gina describes artificial deadlines, sprints, write-ups and show-and-tells as practical levers.
- •Kieran says founders are excited to get back to shipping after the COVID ballooning of companies
- •The faster you ship, the more you learn, and founders set the tempo
- •Gina suggests external or artificial deadlines that can't be pushed
- •Two-week sprints, A/B testing cadence and retrospectives support learning rather than spray-and-pray
- •Duolingo wrote up each experiment's hypotheses beforehand and reviewed results two weeks later
- 28:13 – 34:50
Experimentation and Big Bets: A Portfolio Approach
The panel discusses whether to experiment broadly or focus on proven channels. Kieran frames growth as a portfolio of bets, Gina says to treat everything as an experiment and rank ideas by ROI, and Bryan reflects on intuition and returning to fundamentals.
- •Kieran compares growth initiatives to an investment portfolio mixing high-confidence and speculative bets
- •Gina says that with enough users everything should be an experiment, but the approach is 'and', not 'either/or'
- •Early-stage startups need big swings because 2% improvements won't get noticed
- •Gina would rank ideas by ROI: difficulty versus expected return
- •Kieran cites HubSpot's acquisition of The Hustle as a big bet driven by conviction
- •Bryan says 'the calm sea never made good sailors' and urges a return to fundamentals
- 34:50 – 40:00
Using AI in Distribution: Cost Savings and the 10% Opportunity
Steph asks how companies are using AI on the distribution side. Gina focuses on reducing work and on AI's novelty for brand memorability, while Kieran separates cost-saving replication from the 10% of new capabilities, such as personalized automation guides at Zapier.
- •Gina sees AI copy as 75–80% of the way there and useful for first drafts and automation, though human emotion is hard to replicate
- •People's current fascination with AI makes them more likely to pay attention and remember a brand
- •Kieran says marketing may shift from channel specialists to generalists who think like product builders
- •About 90% of use is replicating existing work, such as outbound sales
- •The 10% is new capability, such as ChatGPT-generated automation guides personalized to a user's role and tech stack
- 40:00 – 44:42
Responding to the AI Tsunami: Fundamentals, Windows and Creativity
Bryan Kim says there is no avoiding the AI wave and the response should return to why the product exists rather than bolting on something shiny. He sees a window where consumers are willing to try new products, and a near-zero cost of creation that makes creativity the new limiting factor.
- •Consumers are currently willing to download and try AI products, creating a window for new entrants
- •Snap avoided calling its lens product augmented reality for four years because users cared about the emotion, not the technology
- •AI should deepen the core reason customers use a product, either by automating rote work or delivering previously cost-prohibitive value
- •Existing companies can deepen their moats, and new companies have a moment of attention
- •With the cost of creation near zero, creativity becomes the new limiting factor
- 44:42 – 49:21
Growth Challenge: What Would You Do with $10K for the a16z Podcast?
Steph challenges the panel to grow the a16z Podcast with $10,000 on the distribution side. Kieran suggests an AI-generated interview with a historical entrepreneur, Bryan proposes a 'Fight us' TikTok stitch campaign, and Gina recommends reaching older business-savvy audiences while warning against gameable metrics.
- •Kieran says podcast growth is mostly grind and proposes an audience-voted AI interview with a deceased entrepreneur
- •Bryan suggests a controversial topic, AI-spliced clips and a 'Fight us. Stitch incoming' call to action, seeded with influencers or founders
- •Gina would target older, business-savvy readers of The Wall Street Journal and The New York Times
- •Gina notes you could game the metric by buying cheap listeners in countries like those in Southeast Asia, but they wouldn't be the core audience
- •If a growth problem can be easily gamed, you may be looking at the wrong thing
