|
GTM stack intelligence, enriched.
|
|
Consolidation continues across the GTM stack. Adyen closed its purchases of Orb and Talon.One, Zoom agreed to buy Common Room, and HubSpot agreed to buy Warmly.
Avoidance of vendor lock-in is a core pillar of Enriched’s view on the future of the GTM stack. That pillar is looking a little shaky.
|
|
|
|
Typically, each story gets its own section. Different names, different events, different dynamics. It usually feels right. Not this week. Two different acquisitions were announced that feel just a bit too similar to separate. So, this week we’re pleased to offer Enriched’s first ever 2-for-1 deal.
|
HubSpot — Agrees to Acquire Warmly, Adding Person-Level Visitor ID and GTM Agents
Warmly announced on June 30th that it has agreed to be acquired by HubSpot. Warmly identifies the anonymous traffic on your site down to the person (you can’t hide), ties it to intent data (we know what you’re thinking), and runs an Inbound Agent and a TAM Agent on top (you’ll be hearing from us). Per Warmly’s own post, contracts, pricing, account teams, product, and integrations are unchanged for now, and the agents keep running while the deal proceeds. The eventual plan is to connect Warmly’s context and agents across the HubSpot platform.
Source: Warmly — Joining HubSpot, June 30, 2026
|
|
|
Zoom — Agrees to Acquire Common Room, Pulling Buyer Intelligence Into Its Revenue Platform
Zoom signed a definitive agreement on July 2nd to acquire Common Room, the AI-native GTM platform that combines a company’s first-party data with real-world buyer signals into a continuously updated view of which accounts are warming up, with AI agents that prioritize them. Zoom plans to fold the technology into Zoom Revenue Accelerator. The deal is expected to close in the coming weeks, no new features have been detailed, and nothing changes for Common Room customers today.
These acquisitions continue the theme of incumbents buying not building. The verdict is still out on whether this is net positive for customers of the megavendors or the smaller tools that have been getting snapped up. Common Room and Warmly customers just traded an independent vendor for a future product feature inside Zoom or HubSpot. Their optionality has been limited, and there is no guarantee that the incumbents can thoughtfully integrate these new capabilities into their existing platforms. So, two courses of action I’d suggest: the questions to raise at renewal are pricing, the integrations you’ve wired in, and how easily your data leaves if you go. Ahead of those conversations, start taking stock of your data, workflows, and outcomes that rely on smaller vendors, and make sure you can replicate or migrate those to alternative tooling if you’re not willing to utilize the acquirer’s eventual version.
Source: GlobeNewswire / Zoom, July 2, 2026
|
|
A closer look at one shift in the stack each issue.
|
|
Stack Deep Dive · Metering
|
The Metering Layer — Adyen Closed Orb and Talon.One
Adyen confirmed on July 1st that it has closed both the Orb and Talon.One acquisitions, with regulatory approvals and closing conditions satisfied. Orb is a usage-metering and rating engine: it counts what a customer consumed and turns it into an invoice line. Talon.One is a promotion and incentive engine. Adyen’s June 11th announcement had set the price at $335 million with a reverse-triangular merger structure.
Usage-based billing infrastructure continues to be a key element in the GTM stack, and the payments layer is absorbing the pricing layer. Stripe did the same when it bought Metronome, which closed in January, and acquiring Orb lets Adyen run the same playbook: count the usage and charge for it in one place.
The consolidation isn’t over either. Salesforce signed its own metering deal, for m3ter, in June.
For RevOps and FP&A teams, metering is critical infrastructure and the backbone of the commercial engine. It underpins product telemetry, pricing and monetization structure, discounting strategy, forecasting, and revenue recognition. Leaving it is not a tool swap but a rebuild of how you measure your product and get paid. Once that layer sits inside a payments provider, your exit gets harder because the meter is now fused to the money movement. If you’re building a usage-billing stack, the decision is less about which metering product is best in isolation and more about whose ecosystem you’re willing to utilize from metering through to billing and treasury.
Sources: Adyen — Closes Talon.One and Orb, closed July 1, 2026; Adyen announcement, June 11, 2026
|
|
|
|
AI for RevOps · Anthropic
|
Anthropic — Claude Sonnet 5 Ships as the New Default, With Introductory Pricing Through August 31
Many of you will have already heard that Fable 5 has been released to the public (see our Quick Hits below for more). We’re taking a slightly different focus. Each issue of Enriched covers more and more agentic capability announcements across the GTM stack. The underlying model of many of those agents just got an upgrade. Anthropic shipped Claude Sonnet 5 on June 30th as the default model for Free and Pro. It has a 1M-token context window, is positioned as the most agentic Sonnet-tier model yet, approaches Opus 4.8 on reasoning, tool use, and coding at better cost-performance at medium effort, and is now the default in Claude Code.
Its introductory pricing expires August 31st. Input runs $2 per million tokens and output $10 per million through then, reverting to $3 and $15 after. Sonnet 5 powers much of the agentic tooling in this issue, so budget agent workloads at the post-August rate rather than the launch one.
Source: Anthropic — Claude Sonnet 5, June 30, 2026
|
|
|
Gong — Lands in Microsoft Marketplace With MACC Purchasing and an M365 Copilot MCP Connection
Gong announced on July 1st that it is now listed in the Microsoft Marketplace and purchasable through Azure Consumption Commitments, so buyers can apply it against money already committed to Microsoft. The integration connects M365 Copilot to Gong’s MCP Server, bringing Gong-captured calls, emails, and meetings across Copilot, Teams, Outlook, and Dynamics 365, with interactions synced into Dynamics 365 as AI summaries, key topics, and next steps.
The MACC purchasing path changes the procurement math: buying Gong as a drawdown on an existing Microsoft commitment shifts the source of the budget drawdown. More importantly, for some of the largest enterprises, many of whom are M365 shops, the MCP connection plugs Gong into the Microsoft surface where many enterprise reps already work. This is interesting timing considering our first Quick Hit.
Source: PR Newswire / Gong, July 1, 2026
|
|
The rest of the week, in brief.
|
|
|
|
Microsoft’s Sales agent in M365 Copilot reaches general availability: Microsoft announced on July 2nd that the Sales agent in M365 Copilot is generally available. It’s a role-based Copilot that brings AI insights, conversational intelligence, and seller workflows into Outlook, Teams, mobile, and Copilot chat, grounded in CRM data, emails, and meetings to flag deal risks and openings. It competes directly with Salesforce Agentforce and Gong, delivered from inside the tools reps already keep open.
Source: Microsoft 365 Copilot Blog, July 2, 2026
|
|
Pipedrive ships a native MCP server on every plan: Pipedrive released a first-party MCP server on June 30th, on all plans, with OAuth, working with ChatGPT, Claude, or any MCP assistant. It supports natural-language search across deals, contacts, orgs, and leads, record creation and updates, lead conversion, pipeline insights, notes-to-records, and multi-step automation, with permissions respected and an audit trail. A mid-market CRM shipping first-party MCP makes MCP table stakes across the whole CRM category, not just the enterprise tier. If you can’t interact in natural language with your CRM data, you might soon be behind.
Source: Pipedrive Newsroom, June 30, 2026
|
|
Anthropic restores Fable 5 and Mythos 5, but Fable 5 comes back only partially: The June 12th export controls were lifted, and Anthropic announced on June 30th that Fable 5 is available again globally from July 1st across Claude.ai, Claude Code, Cowork, and cloud providers. The restoration is partial: Fable 5 is included for up to 50% of weekly usage limits through July 7th, after which standard subscriptions lose the model and access moves to paid usage credits. Mythos 5 returned only to approved US organizations under Project Glasswing. If you rebuilt around Fable 5’s return (that would have been quick!), its full access ends July 7th.
Source: Anthropic — Fable 5 / Mythos 5, June 30, 2026
|
|
Clay adds BigQuery import to Audiences: In its Product Roundup published June 29th, Clay added BigQuery Import to Audiences (pull contacts and companies in via SQL), plus Salesforce Activities in Audiences, Function Tags, and inline schedule presets. With BigQuery Import, Clay now pulls from your warehouse directly rather than only receiving data pushed to it, a stronger data-gravity position than the changelog line implies and a follow-on to last issue’s Audiences launch.
Source: Clay — Product Roundup, June 29, 2026
|
|
|
👁️ Noticed
Salesforce signed an agreement to buy the metering vendor m3ter in June. By this week, much of the web reported the deal as closed on July 1st. There is no primary source for that close. Salesforce’s own page still says “signs definitive agreement,” and its guidance is that it expects to close by the end of July. The July 1st date appears to have originated in an AI-search summary and spread across aggregators from there. Adyen closed its metering deal this week. Salesforce did not.
|
|
|
|
|
In a world being eaten by software, proprietary data and workflows are the only moat. There’s no bigger moat than being the infrastructure that connects the meter to the billing engine in a simple, streamlined architecture. You can rent an LLM by the token, but you can’t rent a meter-to-revenue workflow, so Adyen and Stripe bought them, and Salesforce is trying to.
Each of these deals raises switching costs for the acquired vendor’s existing customers, and none of those customers chose it. Metering is already the hardest system to leave. Your pricing rules and invoice history accrete inside it, and once it’s fused to a payments provider, leaving means going beyond swapping a tool: it forces a re-architecture of the way you measure and collect payment.
It is a single but poignant example of Enriched’s vendor lock-in lens. In last week’s read, I spoke about the age of abundance. At some point, as the capabilities of frontier models grew, you could scroll through LinkedIn and read any number of posts proclaiming that vibe-coding custom software was easier, cheaper, and better than continuing to renew with incumbent vendors. As most things do, the pendulum retreated from the extreme. Yet it left us with a different, although similar, question: use abundance-mindset vendors who open up their data and capabilities via MCP and APIs, or use vendors like Stripe or Adyen which provide a streamlined platform for complex workflows such as metering and billing?
In my view, we’re firmly out of the build vs. buy debate for a significant amount of use cases. You’re not vibe-coding your way to a NetSuite or Metronome clone. But I believe the ‘abundance’ vs. ‘walled-garden’ debate is still very much alive. Just be aware that you may be one acquisition announcement away from the decision being made for you.
See you next week. — Andrew
|
|
|
|
|
|
