Key Moments
Kevin Systrom: Instagram | Lex Fridman Podcast #243
Lex FridmanKey Moments
Instagram's co-founder Kevin Systrom on its origin, growth challenges, sale to Facebook, and future of social media.
Key Insights
Instagram originated from a failed check-in app (Burbn) that pivoted to photo sharing after user data showed photos were the most loved feature.
Early Instagram succeeded by embracing imperfections with filters (like X-Pro II), optimizing upload latency, and fostering a community for both friends and shared interests.
Systrom emphasizes that critical feedback on products often comes from data analysis of user behavior, as direct feedback is often unreliable.
Scaling a startup requires focusing on product-market fit first, then hiring experienced people, and rigorously writing tests for maintainability.
The decision to sell Instagram to Facebook for $1 billion was driven by the opportunity to leverage Facebook's resources for rapid scaling and de-risk the company for its 13 employees.
Future social networks may involve less 'social' interaction, focusing more on algorithm-driven content discovery based on individual interests rather than social ties, akin to TikTok's approach.
THE UNEXPECTED ORIGINS OF INSTAGRAM
Instagram, co-founded by Kevin Systrom, initially emerged from a location-based check-in app called Burbn in 2010. During a period when check-in apps like Foursquare were popular, Systrom and his co-founder Mike Krieger sought to build a better version. However, Burbn failed to gain traction due to a lack of differentiation. This led to a crucial pivot where they analyzed user engagement, discovering that photos were the most beloved feature. They then decided to strip away most other functionalities and focus exclusively on photo sharing, leading to Instagram's creation as a mobile-first, visual social network.
THE GENIUS BEHIND INSTAGRAM'S EARLY SUCCESS
Instagram's rapid growth was attributed to its ability to address key user pain points in mobile photography. Early smartphone cameras produced low-quality images, so Systrom developed filters (like X-Pro II) that aesthetically enhanced these imperfections, making photos more 'artistic' and shareable. The app also cleverly masked slow upload times by processing photos in the background while users captioned them, making the experience feel lightning-fast. The initial decision to use square photos was a result of processing efficiency. This focus on product market fit, even with technical limitations, proved incredibly effective.
DATA-DRIVEN PRODUCT DEVELOPMENT AND SCALABILITY
Systrom highlights the importance of data in product development, asserting that "data doesn't lie" when it comes to user behavior. Unlike subjective user feedback, usage patterns accurately reveal what features resonate the most. For Instagram, this iterative, data-backed approach allowed them to quickly adapt and find product-market fit. Scaling the company, especially during its rapid growth to 50 million users with a lean team of 13, presented significant technical challenges. They eventually migrated to AWS from an in-house server and emphasized writing rigorous tests from the beginning, which allowed them to iterate quickly and confidently without breaking the product.
THE $1 BILLION ACQUISITION BY FACEBOOK
In April 2012, Facebook acquired Instagram for $1 billion. Systrom describes the high-stakes negotiations and the prevailing market sentiment where a $500 million valuation was considered audacious. The decision to sell was a calculated one, aimed at leveraging Facebook's vast resources and talent to accelerate Instagram's growth and de-risk the future for its small team. Despite the immense financial success, Systrom reflects on the initial post-acquisition period as a personal challenge, marked by public scrutiny and a sense of 'we have arrived syndrome,' where the question of 'what's next' became prominent.
THE EVOLVING SOCIAL MEDIA LANDSCAPE AND 'JOBS TO BE DONE'
Systrom theorizes that successful products adhere to a core 'job to be done' for users. Instagram's job was to allow visual sharing of life and foster connection, a principle that guided the development of features like Stories. He critiques Facebook's tendency to chase "shiny objects" by copying competitor features without fully understanding the underlying user job, leading to product confusion. Systrom believes that new social networks will continue to emerge by identifying unmet needs or structural weaknesses in existing platforms, much like Snapchat and TikTok did by focusing on ephemeral content and algorithmic discovery, respectively.
THE FUTURE OF SOCIAL NETWORKS: LESS SOCIAL, MORE DISCOVERY
Systrom presents a provocative thesis: the future of social networks will be "less social." He argues that the current model, which heavily relies on content shared by one's social ties, often leads to divisive content and manipulation. Instead, he envisions a shift towards algorithms that prioritize content discovery based on genuine interest, irrespective of who shared it, similar to TikTok's 'For You' page. This approach, rooted in advanced machine learning and recommender systems, would focus on matching users with diverse, entertaining content, making the platform less prone to the negative aspects of direct social connections.
REINVENTION AND THE PHILOSOPHY OF WORK
Systrom emphasizes continuous learning and reinvention, exemplified by his deep dive into machine learning post-Instagram. He advocates for choosing work that is inherently enjoyable, even when it's difficult, rather than solely chasing wealth or fame. He stresses the importance of hard work in achieving great things, drawing parallels to athletic training where rest is a strategic component for long-term performance, not an end in itself. He also highlights the need for leaders to communicate honestly about the demands of work and for individuals to make mindful choices about their career paths.
LEADERSHIP, VULNERABILITY, AND PUBLIC TRUST
Reflecting on public perception, Systrom suggests that a leader's public image is often influenced by how their product makes people feel. He notes that leaders of social networks face unique challenges due to the emotional and sometimes negative nature of user experiences. He stresses the importance of 'bedside manner,' vulnerability, and accountability in building trust. Systrom believes that for companies like Facebook, addressing criticisms through genuine action and acknowledging imperfections is more crucial than simply denying fault, especially in light of recent whistleblower revelations and public distrust.
MACHINE LEARNING'S POTENTIAL BEYOND SOCIAL MEDIA
Beyond social networks, Systrom sees immense potential for machine learning, especially reinforcement learning, in addressing global challenges like climate change through optimization of energy consumption, planning, and logistics. He believes that future leaders, having grown up with exposure to machine learning, will be better equipped to apply this technology to various industries, leading to significant efficiency gains and innovative solutions. However, he acknowledges the difficulty in implementing and scaling reinforcement learning effectively, citing its complexity compared to more established machine learning paradigms.
ADVICE FOR ASPIRING ENTREPRENEURS
For aspiring entrepreneurs, Systrom offers practical advice centered on passion, aptitude, and market need. He suggests striving to build something one genuinely loves, even if it fails, as the learning experience is invaluable. The 'three circles' framework — what you're good at, what you're passionate about, and what the world needs — guides decision-making. He also emphasizes careful consideration of funding, viewing venture capitalists as team members whose alignment is crucial. Ultimately, Systrom encourages deep personal investment ('burning bridges') and a commitment to purpose-driven work over the fleeting pursuit of fame or wealth.
Mentioned in This Episode
●Products
●Software & Apps
●Companies
●Organizations
●Books
●Concepts
●People Referenced
Common Questions
Instagram originated from a failed check-in app called Burbn in 2010. Facing low user engagement, the founders analyzed user data and realized people primarily enjoyed posting photos, leading them to pivot to a dedicated photo-sharing app focused on in-the-moment visual updates.
Topics
Mentioned in this video
The podcast hosting this conversation with Kevin Systrom.
A film about a man who discovers his life is a reality television show, used as an analogy for the awakening that occurs when one achieves superficial success and realizes there's more to life.
A classic computer game that Kevin Systrom enjoyed playing while growing up in Massachusetts, indicating an early interest in technology.
A popular documentary that explored the negative impacts of social media, influencing public perception of platforms like Facebook.
An early social news aggregator, mentioned as an example of a platform that distributed interesting content to users, much like modern social networks.
A social networking application that was an early competitor to Instagram, notable for reportedly turning down a large acquisition offer from Google and fading into obscurity.
The programming language chosen for Instagram's backend, before its mainstream popularity in machine learning.
An in-memory data structure store that Instagram adopted early, which helped solve many scaling problems.
A check-in app that was popular around 2010, mentioned as a competitor that Burbn initially aimed to improve upon.
Another check-in app from 2010, part of the trend that Burbn aimed to address before pivoting.
The very first filter created for Instagram by Kevin Systrom while on a trip to Todos Santos, Mexico, designed to make imperfect photos look artistically 'crappy'.
A web mapping service, used as an example of a system where machine learning could optimize routes for energy efficiency, though users often prioritize speed.
A discovery engine tool that distributed interesting content to users, acting as an early form of algorithmic content matchmaking.
A fitness tracking and planning software used by cyclists and athletes to optimize training and recovery, mentioned for its data-driven approach to long-term performance.
The precursor company to Instagram, which was a check-in app before pivoting to focus on photo-sharing.
Google's email service, where Kevin Systrom learned about optimizing user experience by pre-loading data in the background, a technique he later applied to Instagram.
The programming language used for the client-side (iPhone-only) of Instagram in its early days.
A Python web framework used for Instagram's backend in its initial development.
A database system chosen for Instagram's early infrastructure, noted for its geo-features, reflecting the initial check-in app concept.
A more common database system at the time, contrasted with Instagram's choice of PostgreSQL, which was considered more 'hipster'.
A cloud computing platform that Instagram eventually adopted after initially using a single physical server, realizing its benefits for scaling.
An AI program by DeepMind that mastered games like Go and Chess through self-play using reinforcement learning, inspiring the potential for similar applications in social networks.
An audio-chat social networking application, discussed as a recent example of a startup that might have turned down large acquisition offers and whose long-term success is still uncertain.
A long-form video platform by Instagram, which Kevin Systrom admits did not perform as well in serving Instagram's core 'job to be done'.
A short-form video feature on Instagram, reflecting the platform's adaptation to trends set by competitors like TikTok.
A GPS navigation software app that provides real-time traffic information, used as an anecdotal example of an algorithm taking a user on a less efficient but data-gathering route.
A feature on Instagram that allows users to share ephemeral photos and videos, designed to fulfill the job of sharing daily life more efficiently than permanent profile posts.
Aerospace manufacturer and space transport services company founded by Elon Musk, mentioned as an example of a new venture funded by capital from previous successes.
A financial services and mobile payment company co-founded by Jack Dorsey, mentioned as an example of a successful venture where Dorsey used his capital to build a new rocket ship.
Kevin Systrom worked in corporate development at Google, and the company is later discussed in the context of its search algorithms and a potential buyer for social networks.
A video-sharing platform that was noted as the last major company to be acquired for a figure close to a billion dollars before Instagram's acquisition by Facebook.
The company that was a precursor to Twitter, where Kevin Systrom worked as an intern and learned about server infrastructure and later worked with Jack Dorsey there.
Mentioned as a large tech company that has not been directly outcompeted in its core business, similar to Google.
Photo-sharing social networking service co-founded by Kevin Systrom, initially started as Burbn, which focused on check-in features before pivoting to photos.
Electric vehicle and clean energy company founded by Elon Musk, discussed in the context of its factory operations during COVID-19 and Musk's public persona.
Technology company known for its integrated ecosystem of products and consistent product launches that delight users, providing a model for sustained consumer loyalty.
Social networking service that acquired Instagram. Kevin Systrom initially turned down an offer to work at Facebook before Instagram's success.
A popular check-in app in 2010, which Burbn initially tried to outcompete, but ultimately Instagram surpassed its model.
A music streaming service, used as an example of a product that successfully overcame latency issues to deliver a smooth user experience by eagerly downloading the next song.
Social media platform that originated from Odeo, and which Kevin Systrom's early work indirectly contributed to; later a competitor to Instagram and Facebook.
E-commerce and cloud computing giant, praised for its product execution and delighting customers with efficient delivery, despite facing labor and environmental criticisms.
A social network that emerged after Instagram, creating a market for ephemeral content sharing, which Instagram later replicated with Stories.
The brand name used by Comcast for its consumer cable television, internet, telephone, and wireless services.
A video-sharing social network, described as a successor that further disrupted the social media landscape by focusing on algorithmically driven content discovery.
An online payments system co-founded by Elon Musk, which he sold early on to fund new ventures like SpaceX and Tesla, demonstrating a strategy of using capital from one success to build another.
Cable and internet service provider, mentioned as an example of a company that often evokes public anger and has undergone rebranding (to Xfinity) due to negative perceptions.
A prominent hedge fund founded by Ray Dalio, known for its unique culture of radical transparency and open feedback, which Kevin Systrom discussed.
An AI research and deployment company, mentioned as a leading entity in machine learning, particularly reinforcement learning.
A visual discovery engine, mentioned hypothetically by Kevin Systrom as a company Elon Musk might not care to acquire, highlighting Musk's focus on larger, more transformative projects.
A business and employment-oriented online service, mentioned as a comparison point for Facebook's early utility in connecting real people with real identities, but in a more formal professional context.
An AI research company, mentioned as a leading entity in machine learning, particularly reinforcement learning.
A social news aggregation, content rating, and discussion website, noted for its strong group functionality and diverse communities.
The parent company of Facebook, which rebranded from Facebook to Meta, compared to Comcast's rebranding to Xfinity to distance itself from negative public perception.
Founder of Amazon, whose public perception is described as one of being impressive due to Amazon's scale, but without the 'sense of wonder' often associated with Elon Musk.
Co-founder and CEO of Facebook, who Kevin Systrom knew from his college days and who eventually led Facebook's acquisition of Instagram.
Author of 'Competing Against Luck' and a proponent of the 'Jobs to be Done' theory, which influenced Kevin Systrom's thinking on product design.
A prominent AI researcher at Facebook (Meta), mentioned as being upset by the public's portrayal of Facebook's efforts in machine learning, and passionate about advancing AI.
Co-founder of Apple, known for his ability to deliver beloved products despite a potentially difficult personality, and for his 'bedside manner' in public communication.
Co-founder of Twitter and Square, admired for his hard work and humility; Kevin Systrom worked with him at Odeo.
CEO of Google and Alphabet, mentioned as a leader who projects trustworthiness and humanity.
Founder of Bridgewater Associates, known for his principles of radical transparency and honesty; Kevin Systrom learned about feedback and conflict resolution from him.
Entrepreneur and CEO of multiple companies (Tesla, SpaceX), cited for his approach to using capital to build new ventures after selling PayPal, and his public persona of inspiring wonder.
The head of the Catholic Church, mentioned as a high-profile user of Instagram, illustrating the product's incredible reach and the unique experiences it opened up.
CEO of Microsoft, mentioned as a leader who projects trustworthiness and humanity.
A venture capital firm that invested in Instagram at a $500 million valuation.
A prominent American newspaper, mentioned as a platform where some tech figures pontificate about AI's future without necessarily having hands-on experience.
A technology news website, mentioned as reporting on Path's decision to turn down a substantial offer from Google.
The university Kevin Systrom attended, where he developed skills that would later contribute to Instagram's creation.
A venture capital firm that invested in Instagram alongside Sequoia Capital.
The mobile device whose advanced camera and network capabilities enabled Instagram's vision of instant photo sharing, especially the iPhone 4.
A video game company whose games were used in seminal machine learning papers to train AI agents, demonstrating early successes of reinforcement learning.
Rectified Linear Unit, an activation function frequently used in neural networks, also cited as a 'general good idea' in network architecture.
An optimization algorithm commonly used in deep learning for training neural networks, mentioned as a 'well-trodden way' in general neural network building.
More from Lex Fridman
View all 546 summaries
311 minJeff Kaplan: World of Warcraft, Overwatch, Blizzard, and Future of Gaming | Lex Fridman Podcast #493
154 minRick Beato: Greatest Guitarists of All Time, History & Future of Music | Lex Fridman Podcast #492
23 minKhabib vs Lex: Training with Khabib | FULL EXCLUSIVE FOOTAGE
196 minOpenClaw: The Viral AI Agent that Broke the Internet - Peter Steinberger | Lex Fridman Podcast #491
Found this useful? Build your knowledge library
Get AI-powered summaries of any YouTube video, podcast, or article in seconds. Save them to your personal pods and access them anytime.
Try Summify free