Summary: On May 28, 2026, Asana closed its first acquisition in eighteen years, buying no-code AI-agent platform StackAI in a deal reported at $75 million. The article presents StackAI’s cross-system execution layer as the missing piece in Asana’s human-agent strategy and says integration speed will determine whether the purchase becomes a growth story.
Eighteen years of never buying a company came to an end on May 28, 2026, when Asana confirmed it had closed its acquisition of StackAI, a no-code platform for building AI agents — a deal TechCrunch reported at $75 million. Fortune observed that Asana deliberately lined the announcement up with its quarterly earnings, presenting the purchase as a headline act in its turnaround story.
The urgency is easy to account for. Asana has shed more than half of its market value since ChatGPT appeared, and co-founder Dustin Moskovitz stepped away from the CEO job in March. The bet from the new leadership is to rebuild the company as “the operating system for human-agent teams” rather than a work-management tool — and StackAI is exactly the piece that ambition lacked.
Terms and Numbers
The fine print matters as much as the check. Tony Rosinol and Bernard Aceituno, StackAI’s two co-founders and both MIT PhDs, are joining Asana as part of the deal; the StackAI name lives on for now, with its technology gradually absorbed into Asana’s AI product line. PitchBook’s figures show total funding of just under $20 million, anchored by a $16 million Series A led by Gradient and backed by Epakon Capital, Lobby VC, LifeX Ventures, and Vercel CEO Guillermo Rauch; the startup came out of Y Combinator’s Winter ‘23 batch. Asana is paying $75 million for a company that raised a fifth of that — in effect, buying an execution layer that would have taken years to build in-house for less than a quarter’s marketing budget.
The Product Behind the Price
StackAI’s pitch is a no-code platform for designing, testing, deploying, and governing custom AI agents. It links into enterprise systems such as Salesforce, Oracle, DocuSign, and AWS and can read and write data across them, so one workflow can carry itself through several systems without a human hand-off. Customers in financial services, healthcare, and professional services put it to work on support, IT requests, and compliance reviews. Rosinol distilled the difference: “General-purpose agents talk; specialized agents act. We bring the cross-system workflow engine.”
The Stack, Layer by Layer
With the deal done, Asana’s human-agent stack falls into three layers. AI Studio handles automation of repetitive processes. AI Teammates are agents that take on assigned work the way staff members do. StackAI fills in the execution layer, carrying workflows out into systems beyond Asana. AI Teammates are the connective piece: they feed project context and approval records built up inside Asana into StackAI workflows, and send the outcomes of those actions back again. CEO Dan Rogers’ official framing: the acquisition “accelerates our roadmap and takes us into the next phase of human-agent work,” putting Asana in position to pursue “agentifying the most complex business processes end-to-end.”
Earnings Filed the Same Afternoon
A takeover needs financial backing, and Asana supplied it that same afternoon. For the first quarter of fiscal 2027, adjusted EPS landed at $0.10, against $0.05 a year earlier and $0.07 expected; revenue came to $205.1 million, about 10% above the prior year and ahead of both the company’s own guidance and the $203.9 million consensus; and adjusted operating margin reached 11.5%, an all-time high for the company. Full-year revenue guidance moved up to $855.5 million to $863.5 million, with StackAI forecast to contribute roughly 50 basis points of the growth. The stock gained more than 3% in late trading when the announcement landed, and Fortune recorded a peak above 13% as the earnings took hold. Record margin, higher guidance, and a tuck-in acquisition arriving in one afternoon — to public-market investors, that combination reads as a management team carrying out a coherent plan.
The Message to Work-Centered SaaS
At stake is a question the whole industry is asking: with OpenAI, Anthropic, and Zapier all headed into agents, what keeps work-centered SaaS defensible? Asana’s response is context — which work was assigned, to whom, and which approvals it passed through all live in its system, and standalone agent tools have no path to that data. Opting to acquire rather than build its own is itself a statement about how the market now prices agent execution layers: for a multibillion-dollar public company, $75 million to buy outright beats two years of internal development. Integration speed is the number to watch next. Should wiring AI Teammates into StackAI workflows stretch beyond a year, the acquisition will shift from growth story to a line item on the balance sheet.