Salesforce launched Agentforce Operations generally available: it converts unstructured process documents (SOPs, diagrams, email chains) into structured digital blueprints that specialized AI agents execute autonomously across disconnected enterprise systems. 30+ out-of-box processes including invoice auditing, vendor onboarding, compliance checks, approval routing. Claimed outcomes: 70% cycle time reduction, 80% elimination of manual data entry. Architecture detail: multi-agent orchestration with discrete task handoffs — not a chatbot overlay. If enterprise buyers adopt at scale, this creates a stickier, higher-value revenue surface than per-seat CRM licenses, and the consumption model means revenue grows with usage rather than headcount.

Salesforce stock fell to approximately $167.80 this week, extending the 2026 decline to roughly 35% from January highs and putting the stock within 5% of BofA’s $160 Underperform target. No fresh negative catalyst — just continued sentiment erosion ahead of the May 28 print. With consensus targets averaging $263–$278 (implying 55–65% upside from here), the gap between institutional models and trading reality is extreme. The dispersion across 49 analyst estimates spans $315 ($160 to $475), with standard deviation of ~$60. That range reflects genuine uncertainty about whether the AI monetization thesis is real. May 28 is the event that collapses the distribution.

Citigroup lowered its Salesforce price target to $188 on May 12 — joining UBS ($185, May 21) and BofA ($160, May 18) in a cluster of pre-earnings cuts. The pattern: no analyst updating CRM ahead of the May 28 print is raising targets. All are cutting or holding. This signals institutional uncertainty about the Q1 print, not a fundamental view change. A strong earnings beat with accelerating Agentforce ARR could trigger a wave of upward revisions within 48 hours of the report — the same dynamic that drove ServiceNow’s 17.6% post-earnings recovery after its own “SaaSpocalypse” scare in April.

Q4 FY2026 (Feb 25) set the comparables for May 28. Agentforce + Data 360 combined ARR exceeded $2.9B: $800M Agentforce ARR (up 169% YoY) + $1.1B Informatica Cloud ARR. 29,000+ Agentforce deals (up 50% QoQ). Accounts in production up ~50% QoQ. 60%+ of bookings from existing customer expansion. FY26 Data 360 ingested 112 trillion records (+114% YoY). These are the sequential metrics investors will map Q1 against. Critical question: can Agentforce ARR sustain triple-digit growth as the denominator grows, or does the rate compress? The answer on May 28 determines whether $2.9B ARR is a floor or a ceiling.

Technical detail behind Agentforce Operations: Salesforce converts existing process documents into “digital blueprints” — structured instruction sets that specialized agents use to execute multi-step workflows across email, ERP, and procurement platforms. 30+ included blueprints span invoice auditing, vendor onboarding, PO rescheduling, compliance routing, and approval chains. Consumption model: enterprises pay per workflow executed, not per seat. This architecture matters strategically: back-office automation is potentially a larger market than CRM seats, and it is not vulnerable to the headcount-compression dynamic that is repricing legacy SaaS. If it lands at scale, Agentforce Operations is the product that justifies the consumption-model pivot.

SaaStr argues that as much as 70% of the traditional SaaS spending slowdown reflects incremental enterprise IT budget flowing to Anthropic, OpenAI, and AI infrastructure vendors — not demand destruction. This reframes the competitive threat for Salesforce: the fight is not just Microsoft Copilot vs. Agentforce, it is Salesforce vs. the entire AI-native stack for the same discretionary budget. The strategic implication: vendors that position as orchestration layers running ON TOP of Anthropic/OpenAI (rather than competing with them) may capture both the AI infrastructure budget AND the CRM seat. Agentforce’s integration with Anthropic Claude and its multi-model approach is evidence Salesforce understands this dynamic.

Summer ’26 goes GA June 15 with three structural upgrades: (1) Multi-Agent Orchestration — agents share context and collaborate as coordinated teams across channels; (2) 50+ specialized IT agents deployable in Slack and Teams out of the box; (3) Tableau MCP — AI agents query Tableau analytics directly via secure open integration. Also GA: Customer Engagement Agent (24/7 lead qualification) and Agentforce Self-Service (10-click setup). The June 15 date means deployments start during Q2 FY2027 — bookings from this release should show up in the Q2 print. Volume of shipped product is the counter-argument to “all hype, no delivery”; the question is whether shipped translates into booked ARR.

Workday Q1 FY2027 (ended Apr 30): total revenue $2.542B (+13.5%), subscription revenue $2.354B (+14.3%), operating income $338M at 13.3% margin vs. $39M (1.8%) a year ago — an 11-point margin expansion in one year. Co-founder Aneel Bhusri (back as CEO) is betting on headcount staying flat: AI productivity gains replace incremental hires. Agentic AI adoption doubling QoQ across 4,000+ clients, with ARR approaching $500M annualized. Best Q1 new ACV growth in five years. The contrast with Salesforce is instructive: Workday is delivering AI revenue traction, margin expansion, AND execution credibility simultaneously. That trifecta is what $CRM needs to show on May 28.

ServiceNow Q1 FY2026: subscription revenues $3.671B (+22% YoY, +19% constant currency). 16 transactions exceeding $5M in net new ACV — up nearly 80% YoY. 630 customers with >$5M ACV. ServiceNow recovered 17.6% after its April earnings-day selloff that briefly triggered “SaaSpocalypse” fears. The recovery signal: strong underlying metrics eventually get recognized even after violent initial reactions. For Salesforce, the read-across is twofold: (1) large enterprises are still signing transformational AI contracts; (2) a bad initial market reaction after earnings is not necessarily terminal. Knowledge 2026’s “autonomous workforce” positioning is landing with enterprise buyers.

HubSpot Q1 FY2026: total revenue $881M (+23% YoY, +18% constant currency). GAAP operating income $27.9M vs. an operating loss of $27.5M in Q1’25 — a $55M swing to profitability. Non-GAAP op income $156.8M. HubSpot is growing faster than ServiceNow, Workday, and Salesforce in the current environment. The outperformance reflects its SMB-first motion and cleaner AI integration (no legacy seat-compression overhang). The contrast is instructive for Salesforce: companies with lighter structural baggage are navigating the AI transition faster. HubSpot’s 23% growth vs. Salesforce’s ~10% is the gap Agentforce is supposed to close by creating new consumption-based revenue.

Jason Lemkin’s landmark post: application software forward P/E collapsed from 84x (2020–22) to 22.7x — now at or below the S&P 500. Never happened before. Not in 2022, not in 2008, not in the dot-com crash. The P/E collapse timeline: 84x → 43x → 33x → 31x → 22.7x over six years. IGV is down 21% YTD, 30% from its Sep 2025 peak. $2T in market cap gone. Thoma Bravo’s Orlando Bravo said publicly that some AI-disrupted software valuations are “very warranted.” Lemkin: “The market is not saying software is temporarily overvalued. It is saying we’re not sure the earnings growth assumptions embedded in even 22x are correct.” Businesses winning are AI infrastructure. Businesses losing are AI-labeled legacy products.

The May 25–29 earnings week is the most consequential in enterprise tech this year. Salesforce (May 28) is the marquee event, but Dell Technologies, Autodesk, and Marvell also report. Collectively these prints will calibrate Q1 enterprise spending before macro headwinds fully materialized. Dell’s AI server demand signals the hardware capex appetite driving hyperscaler build-out. Autodesk’s ARR growth tests design-software seat dynamics in an AI environment. Marvell’s data center commentary contextualizes hyperscaler AI infra investment. And Salesforce’s print will either validate the AI-in-enterprise thesis for the rest of FY2027 — or confirm the bears. Expect a volatile close to the week.