01 · EXECUTIVE SUMMARY
Visible tickets, invisible workflow

The New Business Partners team at RWE processed a high volume of onboarding requests through Jira, passing through IT, Credit, Legal, Accounting, and other support functions before a counterparty became Ready to Trade. Every ticket was visible individually — but nobody could see workflow health at the team level: which requests were stuck, whose workload was piling up, or where nomination times had drifted past target.

I designed and built the JQL queries and native Jira dashboards that surfaced this, covering nomination turnaround, workload distribution, and stalled tickets across 10+ teams. The highest-impact KPI turned out not to be a speed metric at all, but one that caught tickets everyone assumed were already finished.

02 · BUSINESS CONTEXT
Detailed data, no aggregate view

Jira held rich operational data — status, workflow stage, assignee, priority, Ready-to-Trade flag, approval status — but it existed at the individual-ticket level. Managers could open any single ticket, but had no standardized way to answer questions like "is this team keeping up?" or "which requests have been sitting untouched the longest?" As backlogs grew and nomination times crept upward — sometimes past 72 hours against an internal 24-hour target — the gap in visibility became an operational risk.

03 · STAKEHOLDERS
Managers across every support function in the chain
  • Operational managers across IT, Credit, Legal, Accounting, and other support functions — the primary dashboard users.
  • New Business Partners team — owned the underlying onboarding workflow the dashboards measured.
  • Individual team members — whose workload distribution became visible for the first time.
04 · BUSINESS PROCESS
From request to Ready to Trade to (eventually) Approved

Requests moved through multiple support functions before reaching Ready-to-Trade status — and then needed to reach a separate, final Approved status. Many teams treated "Ready to Trade" as the finish line and didn't always complete the remaining administrative step, leaving a gap between "functionally done" and "formally closed."

05 · BUSINESS QUESTIONS
What managers actually needed to know day to day
  • Are nominations landing within the internal 24-hour target, by support function?
  • Is workload distributed evenly across team members, or concentrated on a few?
  • Which tickets have been sitting in the same workflow stage the longest?
  • How many tickets are Ready to Trade but not yet formally Approved?
06 · DATA UNDERSTANDING
What Jira already had — just not surfaced

Fields available per ticket: status, workflow stage, created date, updated date, assignee, support function, priority, Ready-to-Trade status, approval status, and resolution information. All the raw material for the KPIs below already existed — the gap was purely in aggregation and presentation, not data collection.

07 · ANALYSIS
JQL first, dashboard second

Every dashboard metric was backed by a custom JQL query — these queries were the actual analytical work; the dashboard was just the presentation layer on top. Requirements came directly from stakeholders across support functions, and dashboards were refined iteratively as feedback came in and reporting needs evolved.

08 · KEY FINDINGS
Three problems hiding behind "everything's in Jira somewhere"
FINDING 1

Nomination times sometimes exceeded 72 hours against an internal 24-hour target — three times over, in a process where trading readiness was time-sensitive.

FINDING 2

Workload was unevenly distributed across team members, invisible without a consolidated view — meaning some people were overloaded while others had headroom.

FINDING 3

"Ready to Trade" wasn't the same as "done." Teams often stopped at Ready-to-Trade status and left the final Approved step outstanding — a gap that was easy to miss ticket-by-ticket but added up to real backlog and compliance exposure.

09 · ROOT CAUSE ANALYSIS
A visibility gap, not a workflow gap

None of these three problems stemmed from Jira lacking the right data — every field needed to detect them already existed. The root cause was aggregation: no one had built the queries and views that would turn ticket-level data into team-level signal. Individual tickets aging in a stage, or sitting at Ready-to-Trade without final approval, were each easy to miss on their own and only became visible once queried in bulk.

10 · RECOMMENDATIONS
Build the aggregate view, then keep it current
  • Nomination turnaround time by support function, measured against the 24-hour target explicitly, not left implicit.
  • Workload distribution across team members, to support proactive rebalancing.
  • Aging analysis — tickets stuck in the same workflow stage, surfaced automatically rather than discovered by chance.
  • A dedicated Ready-to-Trade-vs-Approved view, specifically because it was the gap most likely to be missed and most costly when it was.
  • Ongoing maintenance — refining JQL queries and KPIs as reporting needs evolved, rather than treating the dashboards as a one-time build.
11 · BUSINESS CASE
Closing the gap between "looks done" and "is done"

The dashboards helped bring nomination turnaround times down toward the 24-hour target from a baseline that had sometimes exceeded 72 hours, reduced workflow backlogs, and improved workload balancing across teams. The Ready-to-Trade-vs-Approved view in particular prompted teams to finish outstanding administrative steps, reducing both backlog and the compliance risk tied to delayed formal approval.

10+
SUPPORT TEAMS COVERED
72h → 24h
NOMINATION TIME, BASELINE vs. TARGET
7
CORE KPIs TRACKED
12 · DASHBOARD
The KPI set, end to end
  • Nomination turnaround time by support function, vs. the 24-hour internal target
  • Open vs. closed request volumes
  • Workload distribution across team members
  • Priority-wise request distribution
  • Tickets pending in each individual workflow stage
  • Overdue operational requests
  • Ready-to-Trade requests awaiting final Approved status
13 · LESSONS LEARNED
The best KPI isn't always the obvious one
  • Speed metrics get attention; state metrics get missed. Everyone would have watched a cycle-time chart — almost no one was watching for tickets stuck in a state they assumed was final.
  • If the data already exists, the project is aggregation, not collection. Every KPI here came from fields Jira already captured — the work was in querying and presenting it, not adding new tracking overhead for teams.
  • Dashboards need maintenance, not just a launch. Reporting needs kept evolving, and the dashboards that stayed useful were the ones refined alongside stakeholder feedback.