The story of online education’s rise reads like venture capital mythology: visionary founders, global problems solved by technology, network effects creating unstoppable platforms. But it’s also a story about how two genuinely brilliant solutions to the same problem ended up competing so directly that, by 2025, both found themselves wounded and forced into an embrace neither had anticipated.
In December 2025, Coursera announced it would acquire Udemy in an all-stock deal valued at $2.5 billion. The announcement surprised few observers. Both companies’ stock prices had been battered by market skepticism about online learning’s future, rising competitive pressure, and an existential threat they’d never fully accounted for: artificial intelligence. What’s remarkable isn’t that they merged, but that it took them fifteen years to realize they might be stronger together than apart.
This is the story of how that happened.
The Boy in the Village
Eren Bali was born in 1984 in Durulova, a small apricot-farming village in Malatya, Turkey, situated in a region destabilized by conflict between Kurdish and Turkish populations. Most talented students from the area fled toward Istanbul or Ankara, seeking opportunity and stability. His parents—both educators—made a different choice. They chose to stay, running a one-room schoolhouse where Eren’s mother taught five different grades simultaneously, her determination to lift her community impossible to miss. His father, banned from teaching after the 1980 military coup due to his political activism, found other ways to contribute to their community’s intellectual life.
Young Eren was restless and curious in a way that the village’s educational resources couldn’t satisfy. Mathematics and science classes simply didn’t go deep enough. The village had no advanced textbooks, no specialized tutors, no pathway forward for a gifted child in a place like this. Teachers did their best with limited resources, but the ceiling was visible and frustratingly low.
But something shifted in 1998 when his family, at considerable expense and no small sacrifice, bought their first computer and subscribed to internet access—for just a few months. It was a transformative moment that Eren has described in interviews as life-changing. Suddenly, online, he found mathematics resources he couldn’t access anywhere else. He taught himself, voraciously consuming information from MIT, Stanford, and other institutions whose courses were beginning to appear online. He devoured what was available, limited only by his own curiosity and the quality of his internet connection. By the time he took Turkey’s brutally competitive university entrance exam—one of the most selective in the world, where hundreds of thousands compete for spots at top universities—he’d scored in the top 0.1% of his cohort.
The experience crystallized something profound in him: geography and circumstance didn’t have to determine destiny. If you had access to good information and passionate teachers, you could transcend your circumstances. This wasn’t an abstract philosophy or motivational platitude—it was lived proof. He was living proof. A kid in a remote village had used the internet to compete with students in Turkey’s major cities.
At Middle East Technical University in Ankara, Eren double-majored in computer engineering and mathematics, where he met Oktay Caglar, another computer engineering student with the same intellectual restlessness and conviction about technology’s potential to democratize learning. In 2007, still in Turkey, they built something called KnowBand—a livestream-based learning platform that attempted to let anyone share expertise with anyone else. The idea was right. The execution was wrong. KnowBand never gained meaningful traction. But the conviction persisted.
After graduating and moving to Silicon Valley in 2009 to work as an engineer at SpeedDate, an online dating company that offered practical experience in startup culture and user acquisition, Eren couldn’t shake the pull of education. The dating startup paid the bills but occupied only part of his attention. He attended the Bay Area Founder Institute in 2010, a program designed to help entrepreneurs validate and develop their business ideas, where he met Gagan Biyani, an American entrepreneur with complementary skills and a track record in startups. The two quickly realized they shared the same obsession: could you create a platform where anyone could teach and anyone could learn? Together, they decided to resurrect the idea in a new form, this time with a clearer vision and complementary skills.
A Radical Bet on the Marketplace
Udemy officially launched in May 2010 with a deceptively simple vision: anyone could teach, anyone could learn. The name was a play on “academy,” suggesting universality and accessibility. But the vision represented something genuinely radical—the democratization of who gets to be a teacher. This wasn’t about vetting instructors through credentials or institutions. This was about trusting the market to sort out good teaching from bad.
The founding team faced brutal, relentless skepticism from the venture capital establishment. Fifty different investors rejected them before they found any believer. “Online education is a fad,” they were told repeatedly. “People won’t pay for courses.” “The economics don’t work.” Venture capitalists, despite their reputation for embracing moonshots and disruption, looked at online education and saw a category that would never move the needle for venture returns.
With no institutional backing, the three founders bootstrapped. They built the product themselves in Eren’s apartment, launched without any external validation, and waited to see if anyone cared. The response was immediate and unexpected. Within months, they had over 1,000 instructors and 10,000 students. The market existed. People did want to learn online, and crucially, they would pay. Eren’s conviction had been vindicated.
In 2011, venture capitalist Ron Lee led a $3 million Series A, and in his investment notes, he described Eren’s entrepreneurial vision as being on par with “Elon Musk and Steve Jobs.” But Udemy’s path forward remained uncertain and contested. Early iterations pushed for live, interactive classes where instructors would teach synchronously to cohorts of students. Students didn’t show up. The model bled money. Engagement was abysmal. So the team made a decisive strategic pivot: they would focus exclusively on pre-recorded, on-demand video courses. This decision proved to be crucial—perhaps the most important strategic choice in the company’s history. It enabled infinite scalability. A course created once could serve millions of learners without the instructor having to repeat their effort. An instructor in San Francisco could reach a student in rural Vietnam, synchronously liberated, at any hour of the day or night.
The marketplace model created powerful incentives for all parties. Instructors who drove their own traffic through their own marketing efforts received 97% of revenue from those sales. Students could pay what they could afford—the course listings displayed prices. Udemy took a smaller cut but removed barriers to entry and enabled growth. The system had built-in fairness that was also capitalist. By 2013, after raising a $12 million Series B round from Menlo Ventures and Lightspeed Venture Partners, Udemy had over a million learners and thousands of instructors. The flywheel was turning, and the momentum was undeniable.
Stanford’s Bet on Prestige
While Eren built Udemy, a different vision of online education was emerging at Stanford. Andrew Ng, one of the world’s leading AI researchers and founder of Google Brain, ran the university’s AI lab. In 2008, he launched Stanford Engineering Everywhere, publishing courses online for free—a radical move at a time when universities guarded course materials like proprietary secrets.
That same year, Ng created an online version of his Machine Learning course (CS229) and opened it to the world: video lectures, problem sets, solutions. The response was staggering. Hundreds of thousands enrolled. Not from Stanford. From everywhere. It was proof of something powerful: the demand for world-class education was vastly larger than any institution could supply.
Ng’s colleague, Daphne Koller, a probabilistic modeling expert with equal credentials, witnessed the same phenomenon. When they offered three Stanford computer science courses free online in August 2011, over 100,000 students enrolled in each. More students than Ng or Koller could teach in a lifetime of traditional instruction.
The insight was clear: there was latent demand at a scale nobody had anticipated. But Ng and Koller made a different bet than Eren. Rather than creating a marketplace, they would build a platform around university partnerships. Coursera would partner with the world’s best institutions. Courses would be rigorously vetted. Credentials would carry institutional weight.
In late 2011, Ng and Koller left Stanford to build Coursera full-time. Coursera launched officially in April 2012 with partnerships including Stanford, Princeton, the University of Pennsylvania, and the University of Michigan. The founding round was $16 million from Kleiner Perkins and New Enterprise Associates—investors who immediately understood the opportunity. They believed in a thesis: that online education represented a multi-billion-dollar market, but that this market would be won by platforms with institutional credibility, not consumer charisma.
The two approaches represented fundamentally different bets on what education should be. Udemy believed in radical accessibility and market-driven quality: anyone could teach, and the best courses would succeed through reputation and learner choice. Coursera believed in institutional rigor: the world’s best universities would provide the content, and credentials from those institutions would carry real weight in the job market. One was built on openness. The other on prestige.
Two Markets, Two Philosophies, One Silent Competition
For the first several years after their launches, Udemy and Coursera grew without directly competing. The distinction between their markets seemed clear and defensible. A person seeking an MBA or professional certificate operated in a different universe from someone trying to learn Python quickly and cheaply on a weekend. Coursera’s revenue came from certificates ($30-$70 each) and later from degree programs that universities offered with Coursera handling the technology and managing the student experience. Udemy’s revenue came from course sales, where prices ranged from $10 to $200, and both instructors and the platform took cuts.
Rick Levin, Yale’s former president, joined Coursera as CEO in 2014, signaling something important about the company’s strategic direction: Coursera saw itself not as a startup disrupting education, but as a new form of university, a modernized institution for the internet age. Levin aggressively pursued enterprise and institutional partnerships. Companies wanted to reskill their workforces. Universities wanted to offer online degrees without the overhead of managing large on-campus populations. Governments wanted to provide training to citizens at scale. The institutional market was vast, sticky, and willing to pay.
Udemy, meanwhile, doubled down on instructor growth and marketplace dynamics. The platform’s strategy was deceptively simple: if you could attract enough instructors, and if a small percentage of courses performed exceptionally well, volume would overcome any unit economics challenges. The law of large numbers would work in your favor. By 2013, Udemy had over a million learners. By 2016, that number had grown to 24 million. The platform had become something like the Amazon of online courses—a marketplace with staggering breadth if not consistent depth.
Both companies raised substantial funding. Udemy reached a $1 billion valuation in 2014, a milestone that validated Eren’s original vision. Coursera raised hundreds of millions across multiple rounds, building what appeared to be an increasingly valuable platform. For nearly a decade, both grew without killing each other because the markets they served, while overlapping, remained distinct enough. A software engineer in San Francisco had different learning needs than a corporate HR manager evaluating training platforms. A student seeking a degree had different needs than someone trying to master Photoshop before freelancing. The distinction allowed both companies to thrive.
The YouTube Question That Neither Platform Fully Resolved
But looming over both platforms, unspoken but never quite answered, was a persistent question: Why didn’t YouTube kill online education?
YouTube, after all, had been offering millions of educational videos since its launch in 2005, all for free, with increasingly sophisticated search and recommendation algorithms. If you needed to understand calculus, you could find a dozen different explanations. If you needed to debug Python code, someone on YouTube had probably solved your exact problem. If you wanted to learn graphic design, Photoshop tutorials were abundant. Why would anyone pay for Udemy or Coursera when YouTube was free and often surprisingly comprehensive?
The answer revealed something important about the difference between information and education, between content and learning. YouTube excels at just-in-time learning: solving an immediate, specific problem right now. The platform is optimized for the learner who has a tactical need and searches for a solution. A fifteen-minute video perfectly serves this use case. YouTube’s search is excellent. Its recommendations are increasingly sophisticated. The barrier to entry is zero.
But YouTube was not and is not optimized for sustained, comprehensive learning. There’s no curriculum. No progression from beginner to advanced. No credentials. No feedback from instructors. No accountability. If you wanted to go from zero knowledge to job-ready proficiency in a skill, YouTube would require you to stitch together dozens of videos, figure out which ones were authoritative and which were garbage, manage your own pacing without external support, and maintain discipline to actually finish something without built-in structure.
Udemy solved this problem by organizing videos into comprehensive courses with intentional progression, quizzes to check understanding, completion certificates, and instructor-student relationships. Coursera added something different: institutional credibility. A certificate from Coursera, earned from a course taught by a Stanford professor or a University of Michigan instructor, meant something in the job market. Employers recognized it. Universities granted credit for it.
So while YouTube was a competitive threat in some abstract sense, it was more accurately a complement than a true competitor. Both Udemy and Coursera could point to YouTube and say: “Yes, that’s available. But here’s what we offer that YouTube doesn’t—structure, credentials, and accountability.” This distinction proved durable and allowed both platforms to grow even as YouTube remained dominant in casual learning. The markets weren’t zero-sum.
The IPO Moment and the Profitability Reckoning
By 2021, both companies had achieved sufficient scale, revenue, and strategic position to justify going public. In October 2021, Udemy held its IPO at a valuation exceeding $3 billion. The opening was strong. Investors were euphoric. Eren Bali’s original vision of democratizing education—reducing barriers to learning, enabling anyone to teach, creating a global marketplace of knowledge—seemed vindicated by market enthusiasm.
Coursera followed just months later in March 2021 at approximately $4 billion valuation. The market was almost universally bullish on online education’s future. The pandemic had accelerated digital adoption. Remote work was here to stay. Upskilling was becoming an economic necessity. The timing seemed perfect.
But going public changes everything in ways that growth-stage startup founders don’t always anticipate. The narrative shifted immediately from “growth at all costs” to “show me the path to profitability.” The questions changed. Wall Street didn’t want to hear about market opportunity. It wanted to know about margins, unit economics, customer lifetime value, and paths to sustainable profitability.
And here, both companies faced deeply uncomfortable truths that they couldn’t spin away.
Coursera had accumulated approximately $66.8 million in net losses on roughly $293 million in revenue. The company was spending $107 million annually on marketing alone—more on customer acquisition than some entire companies generated in revenue. Despite years of operation and consistent revenue growth, a clear path to profitability remained elusive. The business required constant investment in course development, instructor partnerships, and customer acquisition. Growth was real, but it came at a cost.
Udemy faced similar challenges, with the added complexity of a fragile instructor ecosystem. While the platform had reached 40 million learners by 2021, profitability remained distant. More pressingly, instructor relationships were becoming increasingly strained. The revenue-sharing model created persistent tension: instructors who drove their own marketing received 97% of revenue from sales they directly generated. But instructors who relied on organic marketplace discovery—Udemy’s own search, recommendations, and promotional email—received only 37% of revenue. Successful instructors began asking themselves a simple, devastating question: Why do I need Udemy at all? If 97% of my sales come from my own marketing, why don’t I just sell courses on my own website and keep everything?
Both companies had also fundamentally underestimated customer acquisition cost in their financial models. Acquiring new learners was expensive and getting more expensive every year. Search costs were climbing as both companies bidded against each other for the same keywords. Marketing efficiency was declining. When you netted out marketing spend against revenue, the unit economics of individual course sales or subscriptions looked considerably less attractive than forward-looking business plans had suggested.
The market, freshly public and armed with quarterly earnings expectations, demanded clarity on the path to profitability. Neither company provided a satisfying answer.
The Competitor Nobody Anticipated
In November 2022, OpenAI released ChatGPT to the public. The release was treated as a technology story at first, interesting to AI researchers and technologists but not necessarily consequential to the broader economy. Within weeks, this assessment proved wildly wrong.
ChatGPT was capable of explaining complex concepts with clarity and nuance. It could provide personalized tutoring across virtually any domain. It could answer highly specific questions instantly. It could write code, debug programs, and explain why code wasn’t working. It was free. It was available 24/7. It was interactive in ways that pre-recorded videos could never be. It improved continuously as you asked follow-up questions.
For platforms built on the premise that online video instruction and credentials were the future of learning, ChatGPT felt like a fundamentally different kind of threat.
The impact on online education was subtle at first, then catastrophic in specific segments. Basic programming courses, especially those targeting beginners, began experiencing revenue declines in 2023 and 2024. Excel tutorials faced new competition from an AI tutor that could explain spreadsheet functions interactively, answer specific questions, and adapt to individual learning styles. The courses suffering most were foundational ones—exactly where platforms like Udemy had built their volume and where network effects created positive feedback.
Chegg, an ed-tech company providing homework help and textbook rentals, experienced the existential threat firsthand. The company laid off 46% of its workforce in 2024 after multiple quarters of catastrophic revenue decline. Executives attributed the decline to two factors: ChatGPT was making their homework help service obsolete, and Google’s new AI-powered search summaries were capturing search traffic that Chegg once dominated. Chegg had seemed invulnerable just eighteen months earlier—millions of users, decades of brand recognition, and an indispensable product. Then a better alternative emerged, and the company’s value proposition evaporated with stunning speed.
It was a cautionary tale that neither Udemy nor Coursera could ignore.
Both companies immediately acknowledged the threat, though carefully and strategically. In SEC filings, Coursera stated with unusual candor: “AI could displace or otherwise adversely impact the demand for online learning solutions, including our offerings.” It was an unusually explicit acknowledgment that the business model itself could be threatened by the very technology everyone was celebrating.
Both companies began aggressively emphasizing AI integration as opportunity, not just threat. Coursera launched “Coursera Coach,” an AI tutor designed to provide feedback and personalized guidance. Udemy rolled out “AI-powered microlearning experiences” designed to provide shorter, more targeted lessons adapted to individual learning patterns. Both companies pursued partnerships with OpenAI and other AI companies, seeking to integrate generative AI into their platforms rather than compete against it.
But beneath these optimistic announcements lay genuine uncertainty about the long-term business model. If AI could teach better than human instructors—more patiently, more adaptively, more interactively—what was the value proposition of an online course taught by a human? If AI could answer any question instantly, why would someone pay for a structured course? If AI could provide personalized tutoring at no cost, what competitive advantage did Coursera’s university partnerships provide?
Market Crisis and the Search for Narrative
The cumulative effect of these pressures became undeniable by late 2025. Both companies were posting revenue growth—growth that, in absolute terms, remained solid and even respectable. But growth rates were slowing. More importantly, the market’s enthusiasm had curdled into skepticism. Udemy’s stock had fallen 35% during 2025 alone. Coursera was down 7%. Both traded well below their post-IPO highs, a decline that reflected not just normal market cycles but fundamental questions about the category’s trajectory and future.
More fundamentally, investors were experiencing what might be called MOOC fatigue. Online education had been hyped since the early 2010s. It had delivered moderate, solid results—millions of learners, meaningful revenue growth, demonstrated product-market fit. But it hadn’t been transformative. It hadn’t created the trillion-dollar company that early believers had imagined. And now it faced new structural headwinds that nobody had fully anticipated.
Consumer growth had plateaued. Both Udemy and Coursera could boast tens of millions of registered users, but converting those users into sustainable, growing revenue was proving harder than any projection had suggested. Enterprise growth remained steadier but faced increasing competition from players with deeper resources and existing enterprise relationships. Microsoft, through LinkedIn Learning, was aggressively pursuing the corporate training market with the full weight of the Microsoft enterprise distribution machine. Google had launched its own career certificates program. Amazon had begun investing in upskilling initiatives. Traditional enterprise learning and development vendors, which had initially been nervous about the MOOC disruption, had mostly adapted and remained relevant.
Both companies needed a new narrative. They needed to show investors a clear path forward that accounted for AI disruption, slowing consumer growth, and increased enterprise competition. That’s when, quietly at first, the merger idea began to emerge from strategic discussions.
The Merger: A Question Answered Pragmatically
Coursera’s acquisition of Udemy for $2.5 billion was pragmatic rather than visionary. Both companies had been losing the narrative battle. Both faced the same fundamental question: How does online education remain relevant in the age of AI?
A merger wouldn’t answer that question, but it addressed several structural problems.
First was redundancy. Udemy and Coursera competed for the same learners, bidding against each other for search visibility, burning capital in overlapping marketing efforts. Coursera projected $115 million in annual cost synergies within 24 months through eliminating this redundancy.
Second was complementarity. Udemy had a massive marketplace of 155,000+ courses and millions of individual instructors. Coursera had partnerships with 250+ universities offering rigorous, accredited content. One platform excelled at breadth; the other at credentials. Together, they could serve a more complete spectrum of learning needs.
As Coursera CEO Greg Hart explained: “We’re at a pivotal moment in which AI is rapidly redefining the skills required for every job. Organizations and individuals need a platform that’s agile and comprehensive. By combining the complementary strengths of Coursera and Udemy, we’ll be in a stronger position to address the global talent transformation opportunity.”
The market’s reaction was mixed. Coursera’s stock jumped 4%. Udemy’s stock rose nearly 22%, reflecting Udemy shareholders’ relief at an exit at a premium to the current price. But the broader market sentiment remained cautious. The combined company was larger but still operating in an uncertain environment.
Two Philosophies Attempting Reconciliation
What the merger truly represents is an attempted reconciliation of two fundamentally different visions of education that have coexisted in tension since 2010.
Eren Bali’s vision, rooted in his experience as a gifted kid in a remote village, was built on radical democratization: anyone could teach, anyone could learn, and the marketplace would ensure quality. Udemy embodied this belief. It said: trust the market. Trust learners to choose good courses. Trust instructors to serve them well.
Andrew Ng and Daphne Koller’s vision was built on the conviction that education required rigor, credentials, and institutional accountability. Coursera embodied this belief. It said: trust the universities. Trust that institutional partnerships ensure quality. Trust that credentials carry real weight.
For fifteen years, these visions operated in different markets, addressing different needs. But as both companies matured, the line between those markets blurred. Corporate training departments wanted both structured credential programs and flexible, just-in-time learning. Individual learners wanted the practical skills of Udemy and the credentialing power of Coursera.
The merger is an attempt to say: you don’t have to choose. You get both. The question is whether a combined organization can hold these philosophies together or whether they’ll ultimately be in tension.
The Unresolved Question
Neither the merger announcement nor subsequent company statements have fully addressed the underlying uncertainty that defines the moment. Can AI and online learning platforms coexist and reinforce each other? Or will generative AI ultimately make platforms like Coursera and Udemy obsolete?
The optimistic case is clear. Generative AI creates unprecedented demand for upskilling and reskilling. Organizations need to understand these tools, work alongside them, and build skills they didn’t need before. Platforms that can help people acquire these skills quickly and credibly will thrive.
The pessimistic case is darker. What if, in two to three years, AI tutoring systems are substantially better than human instructors at explaining concepts? What if AI can assess your learning gaps, adapt to your learning style in real-time, and provide perfectly personalized instruction? What role does Udemy or Coursera play in that world?
Both companies acknowledge this uncertainty in their filings, but they don’t resolve it. The merger doesn’t eliminate the threat. It positions them to better invest in AI capabilities and maintain market position whatever the future holds.
What Remains Uncertain
The story of Udemy and Coursera isn’t finished. It’s simply entered a new chapter, one that will be defined by how the combined organization navigates integration challenges, AI disruption, evolving customer expectations, and a market that remains deeply skeptical about online learning’s long-term viability.
What’s certain is that the optimistic narrative of the early 2010s—that technology would democratize education, that geography and circumstance would no longer determine destiny, that the internet would flatten the educational landscape—has been tested and substantially complicated by realities both founders didn’t anticipate. Capital intensity is higher than expected. Profitability is harder than assumed. Customer acquisition cost is steeper than modeled. Instructor retention is more fragile than predicted. And artificial intelligence, rather than being a tool that online education platforms could control and harness, may be a force that fundamentally restructures the entire category.
The merger represents an acknowledgment of these limits. Two companies built on visionary premises are now focused on practical survival and strategic adaptation. Whether that’s wisdom or capitulation will become clear in the years ahead.
The boy in the Turkish village who taught himself mathematics online using early internet resources imagined that the internet would transform education. He was profoundly right. But the transformation proved more complicated, more contested, and more uncertain than anyone anticipated. The Coursera-Udemy merger is a testament to the complicated reality of how visionary ideas encounter the constraints of building sustainable businesses in rapidly changing technological landscapes. It’s a reminder that good ideas are necessary but not sufficient. Execution, timing, market forces, and unexpected technological disruption all matter enormously. And sometimes, the best way forward is not through continued competition, but through admission that combining complementary strengths might be stronger than standing alone.
