The last Stack Movements closed on June 6 asking whether the foundation-model layer would absorb the GTM application layer from below. Four weeks later the pressure came from the opposite direction: the platforms went shopping. Three acquisitions in eighteen days took out three of the best-known independents in the signals-and-agents layer — and not one of them was priced like a distressed sale. This edition covers roughly June 7 through July 4.

Three acquisitions in eighteen days

June 15 — Salesforce signs a definitive agreement to acquire Fin for roughly $3.6B. Fin is the customer-agent platform formerly known as Intercom, and it claims around 76% autonomous resolution of support requests on its purpose-built Apex model. It joins the Agentforce portfolio, with close expected in Salesforce’s Q4 FY27 (announcement). This is a support-agent deal on paper, but the number matters to everyone in GTM: it is the price a distribution owner will pay for an agent that demonstrably works.

June 30 — HubSpot acquires Warmly (terms undisclosed). Warmly’s person-level website-visitor identification and its Inbound and TAM agents fold directly into HubSpot’s CRM and agent lineup (Warmly’s announcement). We have updated the Warmly profile to Acquired and revised our RB2B vs Warmly decision rules accordingly: if you run Salesforce or any non-HubSpot CRM, treat Warmly’s standalone roadmap as uncertain.

July 2 — Zoom signs a definitive agreement to acquire Common Room (terms undisclosed), merging it into Zoom Revenue Accelerator to form what Zoom is explicitly calling a unified revenue AI platform (Zoom newsroom). Common Room brought reference customers including Atlassian, Anthropic, Notion, Okta, and Snowflake. The strategic surprise is not the deal — it is the buyer. Zoom is now a GTM-platform contender, and we will treat it as one.

Add March’s Pocus acquisition by Apollo and the pattern is unmistakable: the independent intent-and-signals layer is being consolidated into platforms, not funded as standalones. We unpack why — and what it means if you are buying in this category — in our companion analysis, The Great Signal Grab.

Distribution owners are paying for proven agents

Salesforce did not pay ~$3.6B for Fin because it lacked an agent strategy — Agentforce crossed $1B in ARR this quarter by the company’s own account. It paid because two years of platform investment taught it what every GTM engineer already knows: getting an agent to actually resolve work at production quality is the hard part, and distribution does not fix it. The buy-vs-build verdict at platform scale is in, and it favors buying the thing that already works. Expect that logic to keep repricing the application layer upward — at least for vendors with real resolution metrics rather than demo reels.

GTM-native rounds: quieter, but not silent

Venture activity in the category slowed to one notable round in the window — but it was a meaningful one. Aligned closed a $60M Series B led by PeakSpan Capital on July 1, taking total raised to $73.8M and consolidating its position as the digital-sales-room category’s breakout. Our funding tracker now counts six events for Q2 2026 and two already for Q3 — with M&A, not rounds, carrying the quarter’s story. Capital totals on the tracker count venture rounds only; acquisitions are tracked as events.

MCP grows up

The Model Context Protocol moved from Anthropic stewardship to the newly formed Agentic AI Foundation under the Linux Foundation, with Anthropic, OpenAI, and Block as founding contributors and a public-server count above 9,400 (announcement). For GTM builders this is the plumbing story of the year: a vendor-neutral protocol for agents to touch CRMs, enrichment, and sequencers is exactly the counterweight to the platform consolidation described above. Whether the platforms embrace it or fence it is the thing to watch through Q3.

What it means for your stack

If you are a HubSpot shop, person-level intent is about to become a native feature — pause any standalone identification purchase until Warmly’s integration ships. If you run Salesforce, the Agentforce portfolio just got a proven support agent; the GTM question is how fast that capability bleeds into sales motions. If you deliberately run an independent stack, your short list for visitor identification just got shorter — RB2B is now the default independent option at the person level — and every signals contract you sign should assume the vendor may have a platform owner within 18 months. Read The Great Signal Grab for the full buyer’s playbook.