Sign in Start free trial
IT

IT Service Management

One service desk for incidents, requests, changes, problems and on-call.

IT service management gives your IT team one place to run the desk: incidents, requests, changes, problems, knowledge and on-call. It shares people, assets, contracts and finance with the rest of  AI, so tickets arrive with context and outcomes flow on to the ledger. Every workflow works with AI switched off;  AI prepares and a person decides.

111 features available 14 planned 11 capability areas
Highlights

What stands out

Risk you can read

Every change shows a scored risk with the reasons, a backtest and the revenue exposed, so the board decides with evidence.

Major incident console with money at risk

The commander sees the timeline, tasks and the revenue per hour and service credit at stake in one view.

Joiner, mover, leaver in one trigger

A people event orders the laptop, accounts and access, and takes it all back when the person leaves.

Purchase to asset in one chain

A purchase request becomes a requisition, then an asset, then a finance record without re-keying.

On-call that pays

Schedules, paging and the allowance run to Time and Payroll from the same record.

Capabilities

Every feature, in one list

Search or filter. Planned items are labelled; everything else is available now.

Incidents and the service desk

Raise, route, work and resolve IT incidents from one agent workspace, with the caller's context on screen.

20 features
  • Full incident lifecycleNew, assigned, in progress, on hold (waiting on the user, a vendor or a change), resolved with a code, closed and reopened, with every step recorded.
  • Impact and urgency priorityA priority matrix of impact against urgency, set per company and per service, so the same rule sets priority every time.
  • Affected service and itemTie a ticket to the service and the configuration item it hurts, and see related incidents on the same item.
  • Agent workspaceOne list across incident types, queues ordered by how close each ticket is to breaching its service level, bulk actions, templates and macros.
  • Every way inTickets arrive from email, the portal, chat, phone and the API; cases that Service creates for IT are linked to the desk automatically.
  • Assignment and routingRoute by group, skills, round robin or location.
  • Duplicates, parents and childrenDetect duplicates, group children under a parent and link a wave of reports to one mass incident.
  • Help while you typeSimilar incidents, known errors and knowledge articles appear on the ticket form so people self-resolve before submitting.
  • Predicted category, group and prioritySuggests category, assignment group and priority from the text, falling back to your rules when the prediction is unsure.
  • Caller context from the people recordThe caller's manager, location, VIP status, devices and recent tickets are shown from Human Capital.
  • VIP and executive handlingMark people whose tickets get a priority boost and visible handling.
  • Inbound email actionsEmail-to-ticket with inbound actions, so a reply updates the ticket it belongs to.
  • Closure surveysSend a satisfaction survey when a ticket closes and track caller satisfaction.
  • Incident analyticsTime to acknowledge, time to resolve, reopen rate, repeat incidents and top services, from the shared reporting module.
  • Caller confirms the fixResolving stops the clock; the caller confirms and rates it, or rejects it and the ticket goes back to work with a fresh resolution clock, or silence closes it after a set window.
  • Targets and escalation you setService level targets per priority, warnings before the deadline and escalation to owners and managers at it; changed targets apply to new tickets while running clocks keep theirs.
  • Honest pausesWaiting on the caller or on a vendor pauses the clock and the ledger names the true reason; a wait is never counted against the team.
  • Smart defaults on openA new ticket fills category, urgency, location and VIP flag from the people record and the chosen service.
  • Services, items and categoriesMaintain the list of services, their configuration items and ticket categories, and create a service inline from the ticket form.
  • One case engine with ServiceIT incidents, requests and tasks run on the same case engine as customer and HR cases, so clocks, queues and reporting are shared.

Major incidents and status communication

Run a serious outage with a named commander, a live timeline and clear messages to the people affected.

12 features
  • Declare and commandDeclare a major incident, name a commander, open a bridge, keep a timeline and tasks, and plan stakeholder updates.
  • Chat room per major incidentA room opens in your chat tool and the right people are invited automatically. Needs the live chat app connected.
  • Status pageAn internal and a public status page with subscribers.
  • Post-incident reviewA review template with timeline, contributing factors and follow-up actions, drafted from the record for a person to edit.
  • Messages to affected peopleTell affected employees and customers what is happening and what is next.
  • Money at riskWhile a major incident is open, the console shows revenue per hour at risk from orders and the service credit accruing; a person can propose that credit to Finance.
  • Customer mapMap a customer contract or subscription to a service so outages show which customers are hit.
  • Declaration criteriaA company sets what makes a major incident (priority, linked incidents, a VIP, a critical service); the criteria advise by default or can be made to refuse.
  • Communication plan by audienceOne row per audience (executives, affected users, technical teams, public) with channel, cadence and owner, and drafts written from the facts; sending is a person's act.
  • Tamper-proof timelineThe major incident timeline is append-only: entries can be added but never edited or deleted.
  • Review with contributing factorsContributing factors by people, process, technology, communication and external, what went well and wrong, lessons, and actions with owners and dates that continue after publishing.
  • Chain from incident to fixFollow one chain from the incident to the major incident, the problem, the known error and the fix change.

Problem management

Find the cause behind repeat incidents and publish workarounds while the fix is built.

6 features
  • Problem recordsOpen a problem from incidents or proactively, record root cause and track status.
  • Known errors and workaroundsA known-error database; a published workaround shows on the incidents it explains and in knowledge.
  • Tasks and the fix changeProblem tasks and a linked fix change; a problem closes only when the fix is verified.
  • Problems from trendsRepeat incidents by service, item and category surface candidate problems by rule.
  • Root cause toolsPlannedFive whys, fishbone diagrams and a review board.
  • States that must be earnedA known error needs a workaround, a resolved problem needs a root cause and fix, and closing needs a verified fix.

Change enablement and release

Plan, assess, approve and prove changes, with risk explained in plain words.

19 features
  • Change typesStandard, normal, emergency, expedited and model-based changes.
  • Lifecycle, plans and collisionsTasks, implementation, test and back-out plans, and warnings when two changes hit the same item or window.
  • Risk assessment with reasonsA scored risk with the factors behind it, a what-would-lower-it view, a backtest against past changes and the revenue at risk.
  • ApprovalsChange board and emergency board, risk-routed, parallel or quorum approval, delegation, and no one approving their own change.
  • Change board meetingsAgenda, voting, minutes and recurring meetings.
  • Change calendarFreeze and blackout windows, conflicts, and a freeze around finance close.
  • Standard change cataloguePre-approved templates that approve themselves when used as designed.
  • Success tracking and audit evidenceOutcomes, post-implementation review, incidents caused by changes, and IT general controls evidence with a separation-of-duties report.
  • Impact through the configuration databaseSee which services and items a change touches by following relationships.
  • ReleasesRelease records, a calendar merged with the change calendar and a go or no-go decision by a person.
  • Pipeline-driven changeLive pipeline connections depend on your tools.
  • Drafted change plans AI drafts change plans and risk answers for a person to edit; it never approves.
  • Standard change modelsPre-approved models are published through workflow, reviewed on a cycle, and their owner never approves their own.
  • Emergency and blackout safeguardsAn emergency change still needs a retrospective approval, and working through a blackout needs a reason and a second person.
  • Honest back-out attestationThe author of a change can never attest their own back-out plan.
  • Eight-factor risk modelRisk weighs eight factors including revenue exposed, with weights as company settings; the score can raise a change's approval route but never lower it.
  • Sealed evidence packsEvidence is sealed with a hash chain that can be verified later and exported to CSV safe for spreadsheets.
  • Deployment gateA pipeline asks whether a deployment may proceed; the answer and any override within a time limit are recorded in the change evidence.
  • Delivery figuresDeployment frequency, failure rate and lead time from pipeline events.

Service catalogue and requests

A catalogue people can actually order from, with approvals and fulfilment across teams.

13 features
  • Catalogue and formsCategories, items, dynamic forms, audiences, entitlements and translations.
  • Fulfilment plansSequential and parallel tasks across teams with dependencies.
  • ApprovalsBy manager, cost-centre owner, role or condition, with auto-approve where policy allows.
  • Joiner, mover and leaver bundlesStarting, moving and leaving a person triggers the right set of requests from Human Capital events.
  • Purchase hand-offA purchase request becomes a requisition in Procurement and the asset is created on receipt.
  • Request service levelsA service level per item and order tracking.
  • Catalogue governanceAn owner, review date and retirement for each item.
  • Automated fulfilmentCreate an account, add to a group, assign a licence or manage a device through identity and device tools. Live connections depend on your tools.
  • Item designerDesign an item with details, form, plan, approval route, access and a live preview; publishing is a workflow decision and the owner never approves it.
  • Manager-chain approval with fallbackApproval follows the requester's management chain from the people record and falls back to a named role when no manager is on file.
  • Order snapshotEach order keeps the answers and a snapshot of the item as it was ordered, so later edits never rewrite history.
  • Tiered automationGranting access runs by itself while removals ask for confirmation by default; each action's tier is a company setting.
  • Run history for people eventsEach joiner, mover or leaver event runs once, replays change nothing and a re-dated hire moves the scheduled run.

Knowledge and the Employee Center

One place for employees to find answers, order things and track their own tickets.

15 features
  • Authoring and reviewVersions, review and publish by a second person, multiple languages and search.
  • Capture from ticketsTurn a resolved incident or problem into an article as you close it.
  • Ownership and expiryAn owner per article, scheduled reviews and expiry.
  • External knowledge sourcesFederate search across other knowledge sites.
  • Deflection analyticsArticle views against tickets raised.
  • Employee CenterOne portal for IT, HR, finance and facilities: search, catalogue, my tickets, approvals and my assets.
  • Virtual agentLive reset depends on identity connections.
  • Chat apps
  • Multi-language portalPortal and agent in several languages including right-to-left.
  • Phone appPlannedRequests, approvals, agent triage and on-call acknowledgement on the one  AI phone app.
  • Agentic resolutionPlannedAn agent triages, runs a governed fix and closes only with the caller's confirmation.
  • Second publisherArticles are published by a second person, and review and expiry windows are company settings.
  • Expired articles hiddenAn article past expiry is hidden from employees and its owner is told when a review is due.
  • Autonomy tiers for the virtual agentPer action the company chooses whether the agent suggests, asks for confirmation or acts; answers are rule-based with citations, so it works with AI off.
  • Article ratings and containmentEmployees rate articles, and a ticket raised after a no answer is recorded to measure how often self-service resolves the question.

Service levels

Clocks that count the right hours and explain themselves.

4 features
  • Service level clocksResponse and resolution clocks with business calendars, pause reasons, breach actions and an explanation of every minute.
  • Service availabilityAvailability from outages and events, target against actual.
  • Internal and vendor agreementsPlannedAgreements between teams and with vendors, tracked against the same clocks.
  • Service level what-ifPlannedReplay a proposed target on past tickets.

On-call and paging

Who answers tonight, how they are reached, and how they are paid for it.

8 features
  • Schedules and escalationRotations by week, day or 12 hours, overrides, cover and swaps, and escalation policies.
  • Paging channelsPush, SMS, voice and chat paging with acknowledge and reassign. Live SMS and voice depend on a connected provider.
  • Move from another on-call toolImport schedules and escalation policies from the tool you use today.
  • On-call allowanceCompute on-call hours and call-outs against a rate card, approve and send to Time and Payroll.
  • Time-zone-safe rotationsHand-overs happen at the same local time across daylight saving changes, and a company can set its own weekend days.
  • Cover and swaps need a personAn override proposed by the assistant or an API key waits until a person confirms it.
  • Automatic escalationUnanswered pages escalate through the policy steps on a clock, with an append-only page log and a record when nobody answers.
  • Exact allowance in any currencyThe allowance counts hours and call-outs in the schedule's own time zone and currency and is never paid twice.

Automation, API and events

Connect IT to the rest of the company and to your tools.

7 features
  • Approval decision pointsRequests, changes, major incidents and standard changes use the shared workflow module.
  • Automation builderBuild automations with connectors to IT tools.
  • Public API and eventsRead incidents, requests, problems, changes, releases and on-call, raise incidents and order from the catalogue, and subscribe to events.
  • EnvironmentsDevelop and test configuration, then promote to production.
  • External agents over MCPOutside AI agents can work IT records with the caller's rights.
  • Reporting without free textReporting datasets and the SQL reporting role never expose notes, titles, plans or resolutions.
  • Events without ticket textEvent payloads carry identifiers and codes, never the ticket text.

Governance, roles and platform

Separation of duties and AI controls built in.

13 features
  • IT personas and separation of dutiesAgent, resolver, change manager, change board, problem manager, asset manager and service owner roles.
  • Licensing by sub-module and seat typeFulfillers and requesters are licensed differently; on-call is included with the core.
  • AI agent governanceA kill switch, a never-do list, per-agent scopes and an activity log.
  • Everything works with AI offEvery IT workflow is tested with AI switched off.
  • Dashboards and report packsReady reports and persona dashboards for agent, desk manager, change manager, problem manager, leadership and configuration owner, with sharing.
  • AI summaries and draftsTicket summaries, resolution notes, review and article drafts, and chat summaries, written from the record.
  • Hub foundationThree IT hubs with sub-hubs, navigation and licence keys.
  • Audit and evidence packsPlannedPackaged evidence for IT service management, information security and control audits.
  • Multi-entity and managed-service modePlannedSeparate queues and service levels per entity with shared agents.
  • Migration from other service desksPlannedBring tickets, items, catalogue, knowledge and service levels across.
  • Thirteen approval decision points and nine rolesDecision points across the three IT hubs and nine least-privilege role templates, with separation-of-duties rules.
  • Per-action AI autonomyA company sets for each assistant action whether it suggests, confirms or acts; the never-do list is fixed, not a setting.
  • NotificationsIn-app notices for major incidents declared, updates due, status posts, pages, and article reviews due or expired.

Service experience extras

Wider service delivery on the same platform.

8 features
  • HR service deliveryCases, confidential handling and letters for people teams.
  • Customer service casesCustomer-facing cases through the Service hub.
  • Walk-up experiencePlannedAppointments, a queue and a tech bar.
  • Service portfolio and healthPlannedService portfolio, health scorecards and an improvement register.
  • Vulnerability prioritisationPlannedImport, rank and remediate exposures.
  • Process miningPlannedFind bottlenecks from ticket history.
  • Demand and project linkPlannedLink demand to projects.
  • Security incident typePlannedA restricted-visibility incident type for security events, linking out to security tools rather than replacing them.
Assistant

What  AI prepares here

It prepares the work. A person confirms it.

  • Summarise an incident, major incident or problem from its record (nothing is saved)
  • Draft a resolution note from the known error, similar incidents and articles found
  • Catch up a latecomer on a major incident
  • Draft a post-incident review or a knowledge article for a person to edit
  • Raise, update, hold and resume incidents, ready for confirmation
  • Order from the catalogue on someone's behalf, ready for confirmation
  • Plan a release and keep a risk reading
  • Draft a change plan and answer risk questions for a person to edit
  • Never: approve or vote on a change, give a release go or no-go, seal evidence, acknowledge a page, publish a known error or review, or send a stakeholder update. Those stay with people.
Inside every hub

Workflow, reporting, integration, migration

The same four building blocks, specific to IT Service Management.

Workflow

Approvals use the shared workflow module with decision points for requests, changes, major incidents and standard changes. Rules can be written in plain language and previewed with sample approvers; no one approves their own request.

  • Risk-routed change approval: low risk goes to the owner, high risk to the change board, emergencies to the emergency board
  • Catalogue request approved by manager, then by the cost-centre owner above a value
  • Standard change that approves itself when used as designed
  • Disposal and reclaim requests routed to the IT manager, with the requester excluded
  • Quorum approval of a change board with delegation when a member is away

Reporting and analytics

Ready IT reports and persona dashboards on the one shared reporting module; every figure respects the viewer's access.

Datasets

IncidentsRequestsProblemsChanges and change riskReleasesOn-call pages and schedules

Dashboards

Service desk agentService desk managerChange managerProblem managerIT at a glance for leadershipConfiguration owner

Integration

Read and write IT records through the public API, subscribe to signed events, and connect pipelines, chat tools, identity and device tools. Events carry identifiers and codes, never ticket text.

/v1 itsm_incidents, itsm_requests, itsm_request_items, itsm_problems, itsm_changes, itsm_change_risks, itsm_change_incidents, itsm_releases, itsm_release_changes, itsm_oncall_pages, itsm_oncall_schedules

Events

itsm.incident.opened, resolved, closed, reopeneditsm.request.ordereditsm.request_item.approval, closeditsm.problem.opened, state_changed, known_erroritsm.change.opened, state_changed, risk_assessed, pir_recorded, evidence_sealeditsm.release.state_changeditsm.oncall.paged, acknowledged

Connectors

OktaGoogle WorkspaceIntuneGitHubGitLabJenkinsEmail

Migration

Packs to bring service desk data across are on the migration roadmap; the on-call importer is already available.

  • ServiceNowplanned
  • Jira Service Managementplanned
  • Freshserviceplanned
  • BMC Helix and Remedyplanned
  • Ivantiplanned
  • Zendesk (support cases)available
  • Opsgenie and PagerDuty schedules and policiesavailable
Mobile

On the phone

Planned: requests, approvals, agent triage and on-call acknowledgement on the single  AI phone app. The web screens are usable on a phone today.

Who uses it

Service desk agentResolverService desk managerChange managerChange board memberProblem managerRelease managerService ownerIT managerEmployee requester

Industries

  • Managed service providers can run several entities with separate queues (planned)
  • Regulated sectors use change evidence and separation-of-duties reports
Questions

Frequently asked

Does it work without AI?

Yes. Every IT workflow is tested with AI switched off; summaries fall back to text written from the record.

Can the assistant approve a change?

No. It prepares drafts; approvals, votes, go or no-go decisions and sending status updates are a person's acts, and API keys are refused for them too.

Can we use it for HR and customer cases too?

Yes, HR service delivery and customer cases run on the same desk and portal.

How does a leaver's kit work?

A leaver event from Human Capital opens take-back tasks for each device and starts the leaver request bundle.

How are changes kept safe?

Risk scoring, collision checks, freeze windows, board approval, no self-approval and sealed evidence.

Can we move from our current on-call tool?

Schedules and escalation policies can be imported.

Is on-call separately licensed?

It is part of the core service management licence.

See IT Service Management on  AI

Talk to us