Most internal IT service desks run on a tool separate from everything else the business uses, and they spend their days on the same routine requests: password resets, access requests, how-do-I questions, the standard incidents. That is high-volume, rules-based work, exactly the kind AI handles well, and exactly the kind that buries a team and keeps it from the work that needs a human. Agentforce for ITSM brings Salesforce’s AI agents to that problem, on the same platform as the rest of your business.
We deploy Agentforce-powered IT service management: employee help desks, incident and request handling, knowledge, and asset and change workflows. The agents deflect the routine tickets they can resolve safely, route the rest intelligently to the right person, and do it inside a single governed system instead of another silo. Because it runs on the same platform as your customer-facing Salesforce, IT stops being an island and shares the same data, identity, and governance as everything else.
The common failures: an AI desk that tries to deflect tickets it cannot actually resolve and just frustrates employees, routing rules that send the hard cases in circles, stale knowledge so the agent gives outdated answers, and yet another tool bolted on beside the CRM with its own login and its own data. We scope deflection to what the agent can genuinely close, design the routing and knowledge so the rest reaches the right human fast, and keep it on one governed platform.
We bring both the Agentforce production experience and the Salesforce platform depth this requires, as a Salesforce Consulting Partner since 2017 and a Select Services Partner in Anthropic’s Claude Partner Network. We treat ticket deflection as a means to a better employee experience and a freed-up IT team, not a vanity metric, and we build ITSM that earns its place on the same platform as the rest of your business.
The engagements we run most often on Agentforce for ITSM, from first implementation through optimization.
An Agentforce-powered front door for IT requests, password resets, access, and how-to questions, resolved or routed automatically.
Structured incident, request, and problem workflows with SLAs, queues, and escalation on Service Cloud.
AI-surfaced knowledge and self-service portals that cut repetitive tickets.
Configuration, asset, and change-management processes integrated with your CMDB and approvals.
Human-in-the-loop controls, audit trails, and dashboards for deflection, resolution time, and CSAT.
Engagements are measured by movement on the numbers that matter. These are the directions of travel we commit to.
Predictable phases. Clear deliverables. No surprises.
One to two working sessions to map your current state, business goals, and gaps. We come out with a written scope and recommendation.
Documented architecture, realistic timeline, and transparent commercial proposal. No surprises and no hidden scope.
Configuration, development, integrations, data migration, and QA, with weekly demos and on-the-fly adjustments.
Training, change management, hypercare, and ongoing optimization. We do not disappear at go-live.
Consultant-level analysis from the consultants delivering the work.
What each of Salesforce's seven named Agentforce agents does, which are available now, the long-horizon runtime and multi-agent orchestration behind them, and how to decide which one to pilot first.
Read the full post →Agentforce IT Service manages incidents as records on the Salesforce platform, with prioritization, routing, SLAs and escalation to human staff. Salesforce describes agents that create, summarize and help resolve incidents from employee conversations, plus automated major incident detection. Scope what agents may close on their own, and keep complex or security-related incidents with qualified IT staff.
Problem records link related incidents so teams can pursue root causes instead of repeatedly fixing symptoms. Change management covers planning, approval, implementation and review, including change advisory board steps and AI-assisted risk assessment. For Salesforce customers, these processes reuse familiar tools such as approval processes, Flow and reporting, which lowers the learning curve for existing admins.
Employees submit requests for things like access, hardware or software through a unified catalog and track them to completion. Fulfillment can call Flow, integrations or agent actions to provision access or open tasks for other teams. Design catalog items around the approvals and data each request truly needs, rather than copying every form from your previous tool.
Salesforce includes a configuration management database with a service graph showing dependencies between infrastructure, applications and services, and describes discovery options for populating it. The CMDB supports impact analysis for incidents and changes. Confirm which discovery methods and integrations are available for your environment, and whether an existing CMDB should remain the source that feeds Salesforce.
Employees can raise issues, search knowledge and check request status conversationally in Slack or Microsoft Teams, as well as through a portal. Agents answer from knowledge articles and hand off to IT staff with the conversation context attached. Channel choice should follow where employees already work, and identity mapping between chat accounts and Salesforce users must be reliable.
Knowledge articles ground agent answers and help employees resolve common issues themselves, so article quality directly shapes deflection. Salesforce also cites a large library of prebuilt connectors to identity, collaboration, security and HR systems. Validate the specific connectors you need, such as your identity provider or device management, because agent actions are only as useful as the systems they reach.
Document how incidents, problems, changes and requests flow today, including categories, priorities, SLAs, assignment groups and approval chains. Export sample records to see which fields are actually used. This inventory shows what to simplify and what must carry over for audit or compliance. It also reveals which integrations, such as monitoring or identity tools, the new desk depends on.
Decide which configuration item classes, relationships and asset data Salesforce will hold, and whether discovery or an existing source populates them. Align categories and priority matrices with reporting needs rather than inheriting every legacy value. Settle how employees map to Salesforce users and contacts, since routing, entitlements and chat identity all depend on that decision.
Configure incident and request management first, so humans can work reliably inside the new desk. Add queues, SLAs, approvals and notifications, then connect Slack or Teams. Once the manual path is stable, introduce agent actions for narrow, well-understood requests, such as status checks or knowledge answers, and measure outcomes before broadening scope.
Choose what history to migrate, such as open tickets, recent closed records, knowledge articles and configuration items, and what to archive in read-only form. Rehearse loads in a sandbox and reconcile record counts. Decide on cutover timing, how open tickets transfer and how long the old tool stays available for reference before its contract ends.
After launch, review agent conversations, handoffs and reopened tickets to find where answers fall short. Update knowledge articles and catalog items based on what employees actually ask. Track resolution time, escalation rates and employee satisfaction, and expand agent scope only where results hold. Assign clear owners for knowledge, catalog and CMDB data quality.
Agentforce IT Service is sold by Salesforce, and packaging for the service desk, agents, CMDB and connectors can differ from Service Cloud licensing for customer support. Confirm with your account executive which capabilities are included and whether IT staff need specific user licenses. Also ask how employee requesters are licensed and how agent usage is metered. Ask whether Slack or Microsoft Teams channels, CMDB discovery and particular connectors require additional products. If you already own Service Cloud, clarify whether existing licenses apply or a separate offering is needed. Get written answers before building your business case, because consumption terms for agents affect long-term cost.
Migration projects often copy every category, status and custom field from the previous ITSM platform, including workarounds nobody remembers requesting. The result imports old complexity into a new system and blocks agents from working cleanly. Start from the processes you want, keep only fields that drive routing, reporting or compliance and archive the rest.
An agent that cannot find a correct article either guesses or escalates everything, and employees quickly stop trusting it. Audit the articles behind your top request types, retire outdated content and assign owners before launch. Measure how often agent answers lead to reopened tickets, and treat those cases as a knowledge backlog.
ITSM rarely stands alone: monitoring alerts create incidents, identity tools fulfill access requests and deployment pipelines relate to changes. Teams that migrate the service desk without rebuilding these connections lose automation they relied on. List every inbound and outbound integration early, confirm Salesforce connector coverage and plan custom integration work where gaps exist.
Routing, entitlements and conversational channels all depend on knowing which employee is asking. If Slack, Teams, the identity provider and Salesforce records do not match reliably, requests land with the wrong owner or lack context. Define a single authoritative source for employee identity and test mapping with contractors, new hires and departed staff.
A Salesforce Health Check reviews your org and gives you a ranked list of what to fix first, with effort estimates. It is the right starting point before an optimization project, a new cloud or an AI initiative.
It is Salesforce’s IT service management offering, built on the Service Cloud platform and announced as generally available in October 2025. It covers incident, problem, change and request management, knowledge and a CMDB with a service graph. AI agents work with employees in Slack, Microsoft Teams and a portal. Product names and packaging evolve, so confirm current scope and naming with Salesforce before planning.
Yes, but treat it as a redesign rather than a data transfer. Typical migration content includes open and recent tickets, knowledge articles, service catalog items, configuration items and assignment groups. Workflows, scripts and integrations usually need rebuilding in Flow, Apex or connectors. Plan a read-only archive for older history, rehearse data loads in a sandbox and agree how open tickets move at cutover.
Sometimes. It fits well when processes are fairly standard, the organization already invests in Salesforce and employees live in Slack or Teams. Keeping an existing tool can make more sense when it is heavily customized, deeply tied to IT operations tooling or recently renewed. An integration pattern is also possible, surfacing IT requests in Salesforce while the established tool remains the system of record.
You define it. Agent topics and actions set what an agent may handle, such as answering from knowledge, checking request status or opening a ticket. Anything outside those boundaries, or anything the employee asks to escalate, should route to a human queue with the conversation summary attached. Start with narrow scopes, review transcripts regularly and widen them only where accuracy holds.
Service desk analysts work in a console similar to Service Cloud’s, so the daily learning curve is usually manageable with focused training. The larger shift falls on whoever administers the platform: queues, Flow, approvals, permission sets and data model changes are standard Salesforce administration. If your Salesforce admins already support other departments, decide early how IT requests for platform changes get prioritized.
The CRM where sales pipeline lives, when it is built right.
The operating system for modern customer service, case, knowledge, voice, and AI agent assist together.
Scheduling, dispatch, mobile work orders and parts for teams that work on site.
Cross-channel marketing, email, mobile, journey, and data-cloud-powered personalization at scale.
Industry CRM for wealth management, banking, insurance, and lending, built for compliance from the start.
Configure-price-quote, billing, subscription management, and quote-to-cash on Salesforce.
Tech Talk
A monthly brief for the people who own Salesforce, AI and revenue technology
What changed in Salesforce and AI this month, and what to do about it.
One email a month. Written by the consultants who deliver the work, not by a marketing team, for the leaders who make the technology decisions.
Consultant analysis, not vendor recaps. One click to leave.
