This is far from being Salesforce’s only foray into AI canvases. For example, its newly announced Agentforce Sales digital workforce brings a portfolio of sales agents directly into the CRM. These agents handle tasks such as prospecting, meeting preparation, pipeline updates and quoting. They draw on Customer 360 data and work across Salesforce and Slack.

This also goes well beyond revenue functions in CRM. Zoho Canvas, for example, changes standard record views into role-based workspaces that gather the fields, filters and actions a team needs on one surface. Zoho’s AI tools now help users design and populate these layouts. For a marketing and customer experience example, Adobe’s “journey canvases” in Customer Journey Analytics and Journey Optimizer give teams a single place to map journeys, analyze paths and fallout and adjust orchestration and experiments.

The same architectural pattern is emerging across other enterprise domains, linking data and decisions in a single workspace rather than separate systems of record. In verticals such as retail and healthcare, similar canvases could sit on top of point-of-sale, inventory or clinical systems. A retail operations canvas might combine store telemetry, staffing data and promotion performance so managers can adjust schedules and markdowns in one place, while a healthcare operations canvas could bring together care team notes, task lists and routing decisions for patient flow across departments.

Crossover Between UC And Revenue Canvases

For enterprises, these examples show how canvases and agents within revenue systems are evolving alongside UC- and productivity-embedded canvases. This trend also raises questions about turf — where sales work should appropriately begin — and about how AI-driven actions are governed across platforms. It also sets the stage for increased crossover between unified communications and revenue canvases.

This shift builds on earlier integration efforts that connected communication channels to CRM. CRM extensions for e-mail vendors have existed for years. Salesforce and Microsoft Dynamics, for instance, offer Outlook and Gmail integrations that automatically log e-mails into opportunity records. What’s different now is that AI agents inside CRM and other revenue platforms can ingest, summarize and act on e-mail and meeting content without manual logging. They can extract needs, budget, timeline and risk signals from unstructured conversation data and turn them into structured opportunity fields. While earlier integrations mostly synchronized metadata, today’s agents can also orchestrate multi-step workflows including diagnostics, approval routing and next-step proposals directly from conversation content.

That said, core CRM platforms still serve as the systems of record for opportunity data, forecasting and compliance. But they sit at the center of a growing ecosystem. Revenue products built alongside or on top of CRM embed AI canvases that analyze conversations, generate intelligence and drive actions before the data reaches the CRM. Revenue intelligence platforms such as Gong and Zoom Revenue Accelerator transcribe and score calls automatically. CPQ tools including Salesforce CPQ and Tacton configure and price deals. Contract lifecycle management systems such as IronClad and DocuSign CLM negotiate and approve terms. Sales enablement platforms like Highspot and Seismic surface content and coaching. So, while CRM still tracks what happened, revenue products increasingly shape what happens next and feed insights back to the CRM.

Given this complexity — which is only compounded by the growth of AI canvases in non-revenue systems — revenue work can start in different places depending on the team and the workflow. It may begin inside the CRM, where a seller uses a revenue canvas to summarize past interactions, get suggested questions for the next call and capture information on needs, budget and timeline that sync directly to the opportunity record. Or it can start inside productivity software. In Slack or Microsoft Teams, an AI workspace can assemble account history, notes and next steps so managers can update stages, request approvals or draft proposals without returning to the CRM. Or it can originate in meeting platforms where a Zoom or Webex call becomes the canvas. AI captures the conversation and creates summaries, scorecards and follow-up tasks that flow into deal-management processes and revenue systems.

It’s worth noting specifically how UC providers are also moving closer to revenue workflows. Zoom Revenue Accelerator analyzes sales calls to generate summaries, automated scorecards and coaching guidance and proposes next-step actions. RingCentral’s AI Conversation Expert, which succeeds its earlier RingSense for Sales offering, uses AI to analyze calls and meetings, surface revenue intelligence and suggest follow-ups that connect to sales and service systems. These offerings effectively turn UC-native canvases into revenue-adjacent work surfaces. They also increase the need for shared governance over how AI influences forecasts, coaching and deal execution.

The different scenarios discussed here lead to practical — and sometimes tricky — questions about workflow origins. Many enterprises must now identify where work should begin when multiple platforms offer AI workspaces. And this is made more difficult given that a sales organization might have several starting points for customer engagement as noted above.

ITSM And Operations As Generative Workspaces

IT service management platforms are undergoing a similar transition, using generative canvases to turn incidents and changes into curated workspaces for agents and operations teams. This also forces IT and operations teams to decide between collaborating via ITSM-native canvases or using traditional ticket queues.

ServiceNow’s Now Assist capability introduces agentic workflows on top of ITSM records, allowing AI to summarize tickets, propose resolution steps and orchestrate multi-step actions with diagnostics, history and execution details that show what the AI did and why.

For complex incidents, an operations team might open an ITSM canvas that aggregates telemetry, prior cases and knowledge articles into a single workspace. AI can propose runbooks, draft communication updates and trigger automations. Still, supervisors retain control through escalation paths and evaluation tools that test whether workflows complete tasks correctly. In many organizations, these environments can serve as shared canvases for NetOps, SecOps and service desks, connecting incident response with broader change and problem management processes.

A Single AI Operating Model Across Canvases

The use cases described here show why leaders should plan for AI canvases to appear in more of their core workflows over the next several planning cycles — and in the bigger picture, enterprises should stop treating AI canvases as isolated tools. Instead, they should govern them under a single operating model that spans CRM, UC, productivity suites and ITSM.

Rather than forcing a single entry point for every workflow, organizations can segment by use case. Sales discovery and account planning may fit interaction-native canvases anchored in calls and meetings; pricing approvals and renewals belong in CRM canvases tied to finance and compliance controls; operations war rooms and incident bridge calls can start in UC canvases that capture real-time context before flowing into ITSM canvases for updates, automation and audit trails.

Across all of them, leaders need clear rules for identity, data access, and permissible AI actions — mapping who can trigger which actions on which records, how AI changes get logged and reviewed and how to reverse them when needed. Consistent guardrails for retention, access and approvals should prevent new revenue and ITSM canvases from drifting into less-governed silos. Enterprises that start with a few critical workflows and apply consistent governance across all canvases should be better positioned as AI moves from experiment to core operating infrastructure.

It’s clear that AI canvases are spreading across the enterprise technology stack from collaboration interfaces into core operations. As canvases embed more deeply into revenue and operations stacks, the next test will be their impact. Do they improve cycle time, win rates and service quality? Or just create traffic jams and confusion about the rules of engagement. I will be monitoring how leaders assess outcomes to understand ROI and intelligently adjust governance as AI takes root in enterprise operations.