Back to News
Market Impact: 0.3

Salesforce Could Be Undervalued if This Acquisition Solves Its Biggest Growth Problem

Artificial IntelligenceCompany FundamentalsCorporate Guidance & OutlookM&A & RestructuringAnalyst Insights

Salesforce shares are down nearly 40% YTD as investors worry its revenue growth remains stuck despite an “agentic AI” push. KeyBanc downgraded CRM from “overweight” to “sector weight,” citing Agentforce not growing as expected and CIO survey data implying Salesforce could be deprioritized in IT budgets next year. Salesforce is responding with Data 360 plus Informatica (to clean/organize data for AI agents) and agreed to acquire Fin for $3.6B (deal expected to close in fiscal Q4 ending Jan 2027), while the stock trades at a forward P/S under 3 and forward P/E of 11.5—valuation support but needs accelerating growth to re-rate.

Analysis

CRM is now a credibility story, not a product story. At this valuation, the market is effectively saying that AI features will not translate into durable budget share unless the company proves it can raise conversion and seat expansion at the same time; that is a high bar in an environment where CIOs are pruning vendors, not adding them. The acquisitions may improve the product stack, but they also increase integration risk and raise the odds of another 2-3 quarters of “work in progress” messaging rather than measurable acceleration.

The cleaner second-order read is that budget dollars are likely to migrate toward vendors that already own workflow control and data plumbing. That is structurally more favorable to MSFT and NOW than to a horizontal suite trying to patch gaps after the fact, while data-centric tools should see better demand for governance and integration layers. If CRM is forced to bundle more AI to defend share, margin leverage can lag even if top-line growth stabilizes.

Near term, the catalyst path is simple: the stock trades on whether the next couple of quarters show any inflection in organic growth, cRPO, or AI monetization. The bearish thesis breaks if management can show that data cleanup is shortening deployment cycles and improving net retention, but absent that, the low multiple can stay low for longer than bulls expect. The consensus may be underestimating how long enterprise AI adoption takes when customer data is fragmented and implementation friction is the bottleneck, not model quality.