SaaS & Salesforce Intelligence Digest

Salesforce closed the week at $170.77, extending a two-week recovery from the June 26 bottom of $146.32 that capped 14 straight down sessions. The S&P Global consensus target sits near $246 across 52 analysts, with 37 of 51 firms still at Buy-equivalent ratings — roughly 44% implied upside that the market refuses to close. The stock is now back above BofA’s $160 bear-floor target, which briefly inverted in June. What changed isn’t the fundamentals — it’s that the tape stopped punishing every AI headline; the July 14 IBM-driven dip was bought within days.

The definitive first-half accounting: Salesforce fell 40.9% while beating earnings expectations each quarter and raising FY27 guidance — a pure multiple-compression story driven by the “SaaS-pocalypse” panic around AI coding agents. Management’s counterpunch is aggressive capital return: a $25 billion accelerated repurchase executed in March inside a $50 billion authorization, cutting shares outstanding by roughly 10%. With a projected fiscal 2030 revenue target of $63 billion (~11% annualized growth), the setup is binary — if the growth holds, the buyback at these prices is historically accretive; if agents erode seats faster than Agentforce monetizes, it’s an expensive floor defense.

Announced July 7 ahead of the AI for Good Global Summit in Geneva, Salesforce will invest $1 billion in Switzerland over five years to accelerate the country’s shift to agentic operations. Read past the diplomacy: this is Salesforce buying national-scale reference architecture for Agentforce in a wealthy, regulation-heavy market — exactly the environment where trust, data residency, and compliance are the purchase blockers KeyBanc’s channel checks keep surfacing. If Agentforce can clear Swiss financial-services and pharma bars, that’s a repeatable enterprise sales asset, not a press release.

The quiet pricing revolution: Agentforce Help Agent and the Customer Service Portal are generally available this month with guided setup measured in minutes and a pay-per-resolution model — Salesforce gets paid when the agent actually closes the ticket. This is the direct answer to the bear thesis that AI agents cannibalize per-seat revenue: instead of defending seats, Salesforce is building the outcome-priced product line itself, mirroring the economics of the Fin acquisition. The strategic risk transfers to execution — resolution-based pricing only works if resolution rates are high enough to beat the seat revenue it replaces.

The Commerce release keeps widening: the OpenAI integration connecting product catalogs directly into ChatGPT hits general availability in July, with Google Search (including AI Mode) and Gemini app integration following through the summer, plus Storefront Next standing up production storefronts in under thirty minutes. The strategic wager is that discovery migrates into third-party AI assistants and Salesforce becomes the commerce backend of record wherever the conversation happens. It’s the clearest example yet of Salesforce building for the post-website world its own bears predict — monetizing the disruption rather than resisting it.

Buried in the H1 wreckage is the number the bulls keep pointing at: Agentforce reached a $3.4 billion annualized run rate by mid-2026, up from an $800 million ARR base at fiscal year-end — growth that would headline any other software company’s year. Combined with the $3.6 billion Fin acquisition (a customer-service AI already profitable on outcomes pricing) and the M3ter consumption-billing deal, Salesforce now owns the full stack for usage-based AI monetization. The disconnect with KeyBanc’s “customers aren’t ready” checks is the central factual dispute in the stock — both can be true if revenue is concentrating in a small set of AI-ready accounts.

The July 14 session was the sector’s ugliest of the month: IBM’s earnings warning suggested enterprise customers are cutting software budgets to fund hardware, and the market sold everything with a subscription model — HubSpot, Workday, and Asana led the decline, with ServiceNow and Salesforce falling alongside. The nuance the tape ignored: ServiceNow simultaneously raised its Now Assist AI contract target to $1.5 billion, direct evidence that incumbents are converting AI into committed contract value. When a budget-environment scare sells off the companies with proven AI attach, that’s dislocation — the kind that either mean-reverts or marks the start of a genuine spending downturn. Q2 earnings season is the referee.

Worth holding against the selloff: HubSpot guides full-year 2026 revenue to $3.700–$3.708 billion, up 18%, with average subscription revenue per customer up 6% to $11,722 as it pushes upmarket into the enterprise platform tier. ServiceNow’s Q1 subscription revenue of $3.67 billion grew 22% year-over-year with full-year guidance near $15.55 billion. These are not the growth rates of an industry being obsoleted — they’re the growth rates of an industry whose equity multiple is being repriced for a risk that hasn’t shown up in the income statement yet. The spread between fundamentals and multiples is the widest it’s been since 2022.

The forecast that reframed the quarter: Gartner estimates up to $234 billion of enterprise application spending is exposed to “agentic arbitrage” between now and 2030 — roughly 20% of enterprise application SaaS spend by that date — as agents substitute for seat-based workflows. For allocators, the actionable part is the boundary: four-fifths of the spend pool is not at risk on these numbers, and exposure is concentrated in workflow-heavy, low-differentiation applications. Expect this figure to appear in every bear deck and every bull rebuttal through year-end; it is now the reference number for sizing the disruption.

Context for the fear trade: Gartner projects enterprise software spending rising 14.7% in 2026 to more than $1.4 trillion, with generative AI as the primary accelerant, and expects 80% of enterprises to have GenAI-enabled applications deployed this year (from under 5% a few years ago). The global SaaS market is projected around $530 billion for 2026, still compounding double-digits. The macro data describes an industry absorbing AI as its next growth layer, not being replaced by it — the equity market is pricing the tail risk, the spending data is pricing the base case. One of them is wrong, and the gap is unusually wide.