Salesforce (CRM) Stock Plunges 30% YTD as AI Threatens Traditional SaaS Revenue Model

Key Highlights

Table of Contents

CRM shares have tumbled approximately 30% in 2026 amid concerns that AI will undermine the traditional SaaS revenue model
Marc Benioff, CEO, rejects pessimism and insists Salesforce’s AI opportunity is at an all-time high
Agent Albert, a forthcoming AI platform, is scheduled to debut before 2026 ends
23,000 customers out of 150,000 total are now using Agentforce, delivering ticket reductions of up to 40% for certain organizations
The company handled 2.4 billion Agentic Work Units during the most recent quarter, representing 57% growth from the prior period

Salesforce (CRM) is experiencing one of its most challenging years. Shares have declined roughly 30% since January, pressured by mounting concerns that artificial intelligence will fundamentally undermine the software-as-a-service business model the company pioneered.

Salesforce, Inc., CRM

The anxiety centers on a simple equation. Salesforce generates revenue primarily through per-user licensing fees, charging organizations based on employee count. As AI enables businesses to accomplish equivalent tasks with smaller teams, demand for user licenses could contract sharply.

Marc Benioff rejects this pessimistic narrative entirely. “People think we have our back against the wall when in fact the opportunity has never been greater,” the CEO stated in a Wall Street Journal interview.

The pain extends across the software industry. The iShares Expanded Tech-Software Sector ETF (IGV) has dropped 20% during the identical timeframe, while the most vulnerable SaaS companies have experienced declines approximately twice as severe as Salesforce’s losses.

The company began its serious AI transformation in early 2023, when Benioff convened approximately 40 senior leaders for an intensive three-day planning session at Salesforce Tower to rebuild the organization’s annual roadmap around artificial intelligence. Weekend strategy meetings continued for several months afterward.

That strategic shift yielded Agentforce, which debuted in late 2024. The platform enables organizations to construct autonomous AI agents capable of executing tasks including service ticket resolution, lead qualification, and IT request management. The technology has been adopted by 23,000 of the company’s 150,000 customer base.

Tangible performance improvements are emerging. Pearson deployed Agentforce agents to address order status inquiries, process refunds, and resolve lost access code issues — boosting the percentage of customer questions answered without human intervention by 40%. PenFed Credit Union achieved a 40% reduction in IT support tickets by implementing an agent to manage password resets and account access problems.

Agentforce Limitations and Challenges

The technology faces meaningful constraints. Pandora’s chief digital officer reported that Agentforce encounters difficulties with ambiguous or sophisticated customer inquiries — such as providing jewelry recommendations when a customer mentions “my wife likes dogs.” Intricate problems continue requiring human expertise.

Initial customer feedback also highlighted significant time investments required for data preparation before AI systems could function effectively. Salesforce addressed this by incorporating a data-integration layer into its technology infrastructure and acquiring specialized companies focused on data management and AI-powered sales tools.

Agent Albert Launch and Revenue Model Evolution

Before 2026 concludes, Salesforce intends to unveil Agent Albert, an advanced AI platform designed to analyze user patterns and execute actions autonomously. Named after Einstein, the company’s mascot, the technology represents three years of internal research and development.

Regarding monetization, Salesforce abandoned its pure seat-based pricing approximately one year ago. The company now operates a hybrid framework — organizations maintain seat licenses while paying per-action fees for Agentforce utilization. A novel metric called Agentic Work Units (AWUs) measures productivity: 2.4 billion AWUs were completed last quarter, reflecting 57% sequential growth.

Benioff also contends that customers cannot simply code their own CRM systems using readily available AI tools. The data security infrastructure, brand protection mechanisms, and regulatory compliance features Salesforce has developed over multiple decades are exceptionally difficult to reproduce, he argued — even with Claude Code or OpenAI’s Codex.

Salesforce has committed over $300 million to Anthropic since 2023. In February, when the companies announced that Claude Cowork would integrate with Salesforce applications, CRM stock surged 4%.

Stifel analysts observed that “CIOs and CTOs prefer a unified platform that integrates agents, actions, data, and workflows” — a strategic advantage Benioff emphasizes as debates over SaaS viability intensify.

Andreessen Horowitz research indicates that business customers with heavy AI usage increased their median Salesforce expenditure by 3% during the past three months.

✨ Limited Time Offer