|
Welcome back! Leaders from Salesforce, Workday and Snowflake tried different tactics this week to show they’re benefiting from AI and will emerge as winners—even though their fast-growing AI products aren’t boosting overall revenue growth, fueling concerns known as the SaaSpocalpyse. (See details of their financial results here, here and here.) Some leaders signaled that AI was prompting layoffs at their firms or among customers, previously a taboo subject. Some also used unusually sharp language in denouncing rivals, perhaps reflecting the way AI products have sparked competition between software firms that didn’t used to compete head-on. Salesforce CEO Marc Benioff turned his company’s earnings conference call with analysts into a kind of corporate marketing event. In what Wall Street analysts said was an unprecedented move, Benioff on Wednesday spent roughly a third of the call interviewing three customers of Agentforce about their experiences with the product. The customers effusively praised Agentforce, saying the product helped identify customers, automate simple tasks like booking hotel guest reservations and “close” sales deals. More notably, Salesforce also unveiled a new metric to measure the value of work its AI agents do. The Agentic Work Unit measures a task completed by an AI agent, like generating a document. The goal of the AWU is to show customers how many tasks they are actually getting done, relative to the “tokens” of AI—the bits of text read or written by a large language model—they’re essentially paying for. In other words, it can help them track how efficiently they’re using Salesforce’s AI product. Executives Patrick Stokes and Robin Washington implied the company could use the task-based metric to figure out how to charge for Agentforce in the future, given that the cost of running the AI itself would come down over time. ‘SaaSquatch’ Rallying Cry “We're still trying to exactly figure out exactly what these [AWU] numbers mean for us,” Benioff told investors. While Salesforce’s overall revenue growth slowed one percentage point to 8% in the January quarter, when excluding revenue from a recent acquisition, Benioff highlighted Agentforce’s impressive growth to $800 million in annual recurring revenue. “If there is a SaaSpocalpyse, I think it might be being eaten by the SaaSquatch because there are a lot of companies using a lot of SaaS because SaaS just got a lot better with agents as a service.” Following the comment, some Salesforce employees began using SaaSquatch—presumably a play on the word Sasquatch, aka Bigfoot—as an internal rallying cry, according to an employee. Benioff also took time to slam software rival ServiceNow, saying Agentforce was helping to liberate companies from the “purgatory of ServiceNow ” (luckily the latter got its own praise from Nvidia’s Jensen Huang). While Benioff is no stranger to criticizing competitors, the comments may reflect how the advent of AI agents has sharpened rivalries throughout the industry, as every enterprise app is effectively selling the same type of AI products, including new AI “superagents.” Meanwhile, Workday co-founder Aneel Bhusri, who last week returned as CEO of the HR app, said in his company's earnings call Tuesday that he’s not happy with the way AI agents from rival software providers were tapping data customers store with Workday. (A Workday customer theoretically could use such agents to get up-to-date employee information for payroll processing or tax filing purposes.) Workday’s ‘Parasites’ Bhusri focused his ire on rivals rather than the Workday customers who are directing these agents, saying the rival software providers are like “parasites” that are getting a “free ride” at Workday’s expense, adding that the company was “going to put an end to that.” He didn’t name the offending companies, but his statement was puzzling considering that Workday already has a model for charging customers when they use AI agents made by other firms, which Gerrit Kazmaier, Workday's product and technology chief, explained in detail after Bhusri’s comments. The ability to charge customers for using AI agents, including those made by other firms, means “there’s a lot of upside” financially, Bhusri said. The comments sounded similar to those of Yamini Rangan, CEO of customer management software provider HubSpot, who told analysts in an earnings call two weeks ago that the company would no longer allow AI agents to access HubSpot customer data for free. “We will monitor it, we will meter it and we will monetize it,” she said. It’s safe to say every enterprise app incumbent is racing to figure out how to deal with potential AI threats from OpenAI- and Anthropic-powered agents that could reduce the apps’ influence, as we reported in detail Tuesday. In any case, some customers might not want enterprise firms to charge them for using AI agents to access their own data. Thomas Wolf, co-founder of Hugging Face, which helps companies store and use AI models, argued that software businesses that allow AI agents to tap their stored data “will be the one[s] winning the hearth of this new generation of AI-native builders.” It isn’t clear yet how the battle over data will impact database providers like Snowflake. But on Snowflake’s earnings call Wednesday, CFO Brian Robins said profit margins from its AI products—which let customers ask questions and get answers from their data in natural language as opposed to coded commands—aren’t as lucrative as its core database product. Still, Snowflake said it’s making up for that by using AI to improve the efficiency of the business. AI Layoffs? For instance, Snowflake is using AI to enhance the way the company predicts how much of its services customers will use, CEO Sridhar Ramaswamy said. There’s layoffs, too. As it highlighted AI efficiency gains, Snowflake on Wednesday said it laid off about 200 staffers during its last fiscal quarter, meaning the company’s headcount was flat in the period. And while Benioff painted a picture of corporate employees working alongside AI, his company’s earnings call showcased how Agentforce can essentially replace human workers outright. One of the customers Benioff interviewed, Jason Lemkin, CEO of SaaStr, which invests in and hosts events for enterprise founders, said the Salesforce AI product helped the company close nearly $3 million in deals while slashing its already-tiny staff. “We went from 15 humans to 2.5 [humans] and 20 agents, okay? That's a lot of change,” he said.
| 
