|
GTM stack intelligence, enriched.
|
|
ZoomInfo beat Q1 estimates and cut its full-year revenue guidance by $62 million in the same earnings call. The stock dropped 36% in after-hours. The explanation: “AI buyer confusion” triggering seat churn. The same week, Outreach shipped ZoomInfo’s signals directly into its AI agents as a May product release. Anthropic signed a global alliance with KPMG and announced a $1.5B in-house enterprise services venture in a span of three weeks.
There’s a renewal conversation embedded somewhere in all of that.
|
|
|
ZoomInfo — 20% Workforce Cut, Guidance Slashed, and a Pricing Model Shift
ZoomInfo’s Q1 2026 results were technically a beat: $310.2 million in revenue, up 1.5% year-over-year. However, none of that matters much when the company cut its full-year guidance by $62 million, announced it is cutting approximately 20% of its workforce, and is closing its Israel facility only 2 years after its expansion. Unsurprisingly, the stock fell 36% in after-hours.
The earnings call language was the tell that this is not just a bid to improve operational execution: “late-quarter macro and AI buyer confusion triggered a purchasing pause, especially in software, leading to seat churn.” Customers are not sure a ZoomInfo seat is worth the price in a GTM stack that increasingly routes enrichment through a warehouse and pulls from multiple providers. The shift to consumption pricing is a bid to alleviate the incongruency of a seat-based model in a vendor-agnostic architecture.
For teams with upcoming ZoomInfo renewals, the restructuring changes the conversation. A vendor cutting headcount and revising guidance downward needs renewals more than it did six months ago. More importantly, know how your team might be able to take advantage of the same tool with a different pricing model. Map your actual dependency first: what breaks without ZoomInfo, what you could replicate through warehouse-connected enrichment, and what signal quality you would lose that you cannot get elsewhere.
Source: investing.com · seekingalpha.com
|
|
|
Outreach — ZoomInfo Signals Added to AI Agents, Legacy Homepage Retiring June 15
Outreach’s May release adds 20+ ZoomInfo signals into Smart Data Enrichment: hiring trends, leadership changes, funding events, and intent signals, now surfacing directly inside the Revenue Agent and Personalization Agent without a separate lookup step. Signal strength scores appear in Account Activity pages where applicable, so the weight behind a signal is visible at the point of action.
Every RevOps practitioner looks to reduce swivel-chairing and integrate actionable insights directly into the workflows that reps already use day-in and day-out. The goal is reduced complexity and, more importantly, reduced change management by integrating intelligence into the existing sequencing layer. Outreach is compressing the gap between signal and execution by building context into the workflow at the moment of action, rather than maintaining it as a separate data product the rep has to go find. HubSpot’s Breeze agents and Salesforce’s Agentforce are moving in the same direction.
The same release sets a hard deprecation date on the legacy 360 homepage: June 15, with all remaining users migrated to the personalized homepage through September 30. If any reporting or workflow in your org depends on the 360 view, June 15 is the action date.
Source: support.outreach.io — May 2026 release notes
|
|
The feature news is worth your attention. The pattern connecting it to the next section may be worth more.
|
|
|
|
AI for RevOps · Anthropic
|
Anthropic — KPMG Alliance and In-House Services Venture Signal a Two-Channel Distribution Push
On May 19, KPMG and Anthropic announced a global alliance embedding Claude into KPMG’s Digital Gateway platform, which serves KPMG’s 276,000-person workforce and its client base. The initial focus is on tax and private equity clients, with KPMG in the U.S. designated as a preferred consultant for deploying Anthropic AI to PE portfolio companies. The two organizations will co-develop Claude-powered products for PE-focused delivery.
That announcement landed two weeks after Anthropic’s May 4 news: a $1.5 billion enterprise AI services venture backed by Blackstone, Hellman & Friedman, Goldman Sachs, and others. That venture embeds Anthropic’s own applied AI engineers directly inside mid-sized companies to redesign workflows around Claude agents, competing explicitly with large consulting firms and systems integrators.
Taken together, it is clear that Anthropic understands the unrealized benefits of AI are not a capability problem, it’s an implementation problem. They are building the distribution infrastructure to land Claude inside organizations through whatever path gets there first. The in-house venture handles direct deployment into mid-market companies. The KPMG alliance handles delivery through an established Big Four SI with existing enterprise relationships at a larger scale. For RevOps teams evaluating Claude API builds who have stalled on the implementation question, a few new channels just opened up to deploy those use cases.
Source: anthropic.com — KPMG · kpmg.com · anthropic.com — enterprise venture
|
|
The rest of the week, in brief:
|
|
|
|
Salesforce MCP Servers: Salesforce-hosted MCP servers are generally available for Enterprise Edition and above. Any MCP-compatible AI client connects via standard OAuth and can access sObject operations, data queries, Tableau analytics, and product APIs, all governed by the org’s existing permission model through the Agentforce Trust Layer. The Data 360 MCP server, which exposes unified customer data from Salesforce Data Cloud through the same interface, is in developer preview. Summer ’26 production rollout continues through June. If you are on Enterprise+ and building anything agentic against Salesforce data, this is the infrastructure to use now.
|
|
Salesforce Flow Orchestration: Starting with Summer ’26, Flow Orchestration runs are included in Enterprise, Performance, Unlimited, and Developer editions at no additional cost. Flow Orchestration handles multi-step automation across Salesforce objects and has been the underlying infrastructure for most complex RevOps workflow builds in Salesforce. The usage caps that caused real design constraints are gone in June.
|
|
Clay in ChatGPT: Clay launched a native integration making enrichment, prospecting, and personalized outreach drafting accessible directly inside a ChatGPT conversation. Apollo made a similar move earlier in May. The pattern worth watching: GTM enrichment tooling is embedding itself into the AI client rather than expecting practitioners to leave the interface to use it. The question is whether “good enough enrichment inside ChatGPT” starts substituting for workflow-level enrichment builds, or just expands top-of-funnel use cases without touching existing tool spend.
|
|
|
👁️ Noticed
Outreach shipped deeper ZoomInfo signal integration the same week ZoomInfo cut 20% of its staff and shifted toward consumption pricing. Outreach’s dependency on ZoomInfo data just became more visible at the moment ZoomInfo’s business model is under the most pressure it has been in years. That is either very good timing for Outreach customers or a dependency worth mapping before your next contract cycle.
|
|
|
|
|
My read on the week:
The ZoomInfo story goes beyond the stock move.
The 36% after-hours drop on a revenue beat is a reaction to why customers are churning: not because they found a better data provider, but because they are not sure whether the category can survive the conventional SaaS per-seat pricing model. That is a structurally different problem from losing to a competitor. You can compete against a better product. Competing against a customer’s uncertainty about whether your pricing model makes sense in a warehouse-first architecture is much harder.
Per-seat data access made sense when the sales motion was to give a rep access to ZoomInfo. It makes less sense when the workflow is to route enrichment through a data warehouse and pull from multiple providers through Clay or a custom built agentic pipeline. ZoomInfo moving to consumption pricing is an acknowledgment that the second motion is real and growing fast enough to affect guidance by $60 million.
The move to a consumption pricing model several years ago may have been a prescient strategic bet, but this time around it is clearly a move made under visible pressure from competitors and market analysts. Every other data provider in the stack, including intent data, contact data, and signal data vendors, has the same reckoning: can you hold a per-seat price when customers are increasingly building around your data rather than through your interface?
For Enriched readers: the implication is not specific to only ZoomInfo. It is about every seat in your stack where the primary value is access to a proprietary dataset. Map the dependency. Ask whether the data is replicable through warehouse-connected enrichment. Ask whether the vendor could justify the price on consumption terms if you asked. The answers will tell you where your next renewal conversation is actually going to go.
See you next week. — Andrew
|
|
|
|
|
|
