IT managers lack real-time visibility — they discover SLA breaches, backlog spikes and MTTR regressions in the monthly meeting, weeks after they happened. Static reports built by hand in spreadsheets are stale the moment they're exported, and every team builds its own so numbers disagree.

Utilyx models IT reporting as an executable workflow: KPIs are aggregated live from the ticket, change and SLA tables via sql-query, dashboards refresh in real time, and periodic PDF reports are auto-generated and distributed.

The workflow is structured as a decision tree with 3 branches (Live dashboard, Periodic PDF report, KPI alert) and 6 steps end-to-end. Each step is executed by a named actor (Dashboard viewer, sql-query, usePDF, email service, KPI engine, IT Manager) and every action is timestamped.

• Step 1 (Start — Dashboard viewer): the IT manager opens the live dashboard. The workflow registers: dashboard_id, viewer_id, filters (date_range, team, priority, sector), refresh_interval (default 60s).

• Step 2 (sql-query): the engine runs a set of KPI aggregation queries against the ticket, change, sla_metrics and ci_assets tables — computing MTTR (avg resolution time by priority), MTBF (mean time between incidents per CI), backlog (open tickets by age bucket), SLA compliance (% within target), first-call resolution rate, and deflection rate. Results are cached with a 60s TTL.

• Step 3 (Routing): three branches — Branch A (live dashboard): KPIs render as widgets (gauges, trend lines, heatmaps) in the dashboard UI, refreshed every 60s; Branch B (periodic PDF report): triggered on schedule (weekly/monthly), the usePDF module generates a report PDF with all KPIs, trend commentary and top 5 incidents/changes of the period; Branch C (KPI alert): the KPI engine detects threshold breaches (MTTR > target, SLA compliance < 95%, backlog > capacity) and fires alerts.

• Step 4A (Dashboard viewer): the manager drills into any widget to see the underlying tickets, applies filters (team, sector, priority), and exports a CSV snapshot if needed. The dashboard supports 7 languages.

• Step 4B (usePDF): the periodic report PDF includes a cover page (period, scope), KPI summary table, trend charts (MTTR/MTBF/backlog over time), top incidents list, top changes list, and an AI-generated commentary paragraph highlighting regressions and improvements. The PDF is attached to the workflow record.

• Step 5B (email): the report PDF is emailed to the stakeholder list (IT manager, CIO, business sponsors) on schedule, in each recipient's preferred language.

• Step 5C (KPI engine / IT Manager): on a KPI alert, the IT manager receives an email with the breached KPI, current value, target, and a drill-down link to the dashboard filtered to the relevant tickets. The alert includes an AI-suggested action (e.g. 'backlog spike on P3 tickets — consider reassigning 2 agents from P4 queue').

Teams that adopt Utilyx report KPI visibility shifting from monthly stale reports to 60-second live dashboards, MTTR regressions caught within an hour instead of at the monthly meeting, and 8h per week of manual report building eliminated by auto-generated PDF reports with AI commentary.

The sql-query aggregation runs against existing ticket and SLA tables — no separate data warehouse required, the KPI alerts fire on threshold breaches so the IT manager acts before the regression spreads, and the multilingual periodic PDF serves global stakeholders without manual translation.