The headline number and what is actually driving it
Coursera reported second quarter 2026 revenue of 298.6 million dollars, a 60 percent increase over the 187.1 million dollars reported in the same quarter a year earlier. CEO Greg Hart called the quarter an important milestone and pointed to a target of at least 85 million dollars in annual run-rate net synergies from the Udemy integration by the end of 2026. The company also raised its full year guidance to a range of 1.220 to 1.245 billion dollars.
The growth is real, but it is inorganic in the specific sense that matters to anyone evaluating whether to buy Coursera stock or Coursera's platform: the Udemy merger closed May 11 and its revenue is included in Q2 results from that date forward. Comparing this quarter to the prior year without adjusting for the acquisition tells you about deal execution, not about whether Coursera's core product is winning more usage from the customers it already had.
The enterprise segment tells a more cautious story
Enterprise revenue grew 118 percent year over year to 140 million dollars, with gross margin holding at a healthy 79.3 percent. On its face, that looks like the strongest number in the release. But enterprise customer count fell 2 percent over the same period, which means the growth came almost entirely from existing large accounts spending more, plus whatever enterprise customers the Udemy merger folded in, not from Coursera winning a broader base of new enterprise logos.
Net dollar retention in enterprise sits at 91 percent. That figure captures expansion and contraction within the existing customer base net of each other, and a number below 100 percent means that, on average, existing enterprise customers are spending less this year than last, even as a handful of larger accounts and the merger inflate the topline. For a buyer evaluating Coursera as a build-vs-buy option for corporate learning, that gap between topline growth and retention health is the number worth pressure-testing in any renewal or new deal conversation.
Why Andrew Ng is betting outside his own company
The same week as the earnings release, Coursera announced a 100 million dollar strategic investment in LearnVector, a new AI-native learning company founded by Andrew Ng, Coursera's own co-founder and board chairman. LearnVector is positioned around personalized, adaptive learning experiences built AI-native from the ground up, rather than AI features layered onto an existing catalog-and-course architecture that predates the current generation of AI models by more than a decade.
That structure is worth sitting with. The person who co-founded Coursera and still chairs its board is putting fresh capital into a separate company built on different architecture, choosing to start the AI-native personalization problem from a blank sheet instead of retrofitting Coursera's existing platform, now bulked up with Udemy's catalog. Reasonable people can debate how much that reflects confidence in Coursera specifically versus a broader read on the market. What it clearly signals is that even insiders with the deepest possible knowledge of Coursera's architecture see enough advantage in starting fresh to justify building a second company rather than simply funding a bigger AI team inside the first one.
What the merger actually solved, and what it did not
The Udemy merger, which closed in May, created a combined catalog spanning Coursera's university and credential-focused content with Udemy's marketplace of practitioner-built courses. For enterprise buyers, that consolidation solves a real, concrete procurement problem: fewer vendor relationships to manage, broader content coverage under a single contract, and the operational efficiency of one integration cycle and one data processing agreement instead of two separate ones running in parallel.
What it does not automatically solve is engagement. Catalog breadth and course completion are different metrics, and the retention numbers suggest existing enterprise customers are not expanding usage in lockstep with the expanded catalog, even as the overall content library nearly doubles in size. Buyers already on Coursera or evaluating the combined platform should ask directly for usage and completion data segmented by legacy Coursera content versus newly folded-in Udemy content, rather than assuming a bigger catalog automatically translates into better learning outcomes for their own workforce.
The build versus buy calculus this creates
For an L&D or CTO organization deciding how to source corporate learning and upskilling, this quarter offers a genuinely useful data point: scale through acquisition is now cheap to observe from the outside, since Coursera's own numbers show exactly what merger-driven growth looks like set against retention health in the same release. A vendor with rising revenue and flat-to-declining engagement per customer is a vendor whose pricing power may be increasing faster than its actual product-market fit, and that gap tends to show up at renewal time as a harder pricing conversation rather than a better product.
That does not mean walk away from Coursera specifically, since a mixed retention picture during an active integration is a normal, survivable phase rather than a red flag on its own. It means the standard enterprise evaluation questions, usage rates, completion rates, net retention by cohort, matter more after a merger than before one, because the topline numbers are temporarily decoupled from the underlying product experience during any integration period. Ask for the cohort data before renewing, not just the aggregate growth figures highlighted in the press release.
What to watch next
Watch whether Coursera's enterprise net retention improves over the next two to three quarters as the Udemy integration matures. A recovery back above 100 percent would indicate the merger is translating into genuine account expansion beyond catalog addition alone. A continued decline would suggest the combined platform is struggling to convert scale into deeper usage, which matters more to a renewing customer weighing next year's budget than headline revenue growth ever will.
Also watch how LearnVector develops and whether Coursera eventually folds its technology back into the core platform, or lets it stay a genuinely separate AI-native product line competing in the same market. Either outcome tells you something about how seriously even the incumbents believe legacy learning platform architecture can be retrofitted for an AI-native experience, and how seriously they believe it instead needs to be rebuilt from scratch by a team unencumbered by the old one.
