August 7 marked a day of diverging fortunes for software companies
According to CNBC, The Wall Street Journal (WSJ), and other outlets, major software companies' stock prices on the New York Stock Exchange on the 7th (local time) moved to extremes depending on their earnings results and AI readiness. Atlassian, the collaboration software specialist, reported fourth-quarter revenue of $1.8 billion, up 28% year-over-year, and turned a profit with net income of $139 million; its stock surged more than 35% in a single day on the 7th, marking its largest gain since its 2015 IPO. Twilio, the cloud-based communications software specialist, posted second-quarter revenue of $1.5 billion, up 22%, driven by growing demand for AI-powered voice and messaging services, and its stock jumped nearly 20%. In contrast, HubSpot, the marketing software specialist, saw its stock fall 19% in a single day following a lowered third-quarter earnings outlook and weak customer acquisition. Datadog, the monitoring software specialist, also dropped 19%, hit by reduced usage from OpenAI, its largest AI customer.
Airtable, the workflow startup, was valued at $12 billion in 2021 but was acquired for less than $1.3 billion—about a tenth of its peak valuation. According to the WSJ, 86% of private investment deal value in the first half of 2026 was concentrated in AI startups, leaving traditional SaaS companies that failed to transform their business models facing a funding crunch and existential crisis.
Vendor explanations and internal usage records are two different sets of data
In an interview with the Financial Times (FT), Atlassian co-CEO Mike Cannon-Brookes said, "I don't accept the argument that AI will reduce demand for existing products like Jira and erode subscription revenue," adding that "customers are already finding useful ways to leverage AI within our platform."
Such explanations from vendor executives are claims about their own products. Companies paying subscription fees can verify these same claims using their own internal data. By counting how many of the contracted seats actually logged in last quarter, and how many transactions were processed through AI features after they were enabled by the vendor, companies are left with concrete numbers they can use as leverage in renewal negotiations. These numbers cannot be replaced by vendor press releases.
For Korean companies, this is a buyer-side issue
If your company is using collaboration, communications, marketing, or monitoring tools made by the companies named in this earnings cycle, this earnings news is less market information than a vendor checklist item. If you have a contract approaching renewal, the priority is to first check the contract terms for what operations would be affected if the vendor's service went down, and in what format accumulated data could be extracted. Since the first thing to be adjusted at a vendor facing funding difficulties may not be pricing but rather the level of support provided, it's advisable to review incident response clauses and data export clauses together at the same time.
There is one thing to check this week. Make a list of software subscriptions coming up for renewal within the next six months, and for each tool, note the number of actively used accounts and the volume of work that tool handles. If there's a line where the numbers can't be filled in, that's the item to sort out before considering the adoption of new AI tools.