EDGAR OWEN Data Analyst | SQL, Python and Operational Reporting Sheffield, United Kingdom | edgar.owen@example.org PROFILE I turn operational data into clear evidence for everyday decisions. I use SQL and Python to investigate patterns, check data quality, and build repeatable analyses, then present the results through focused reports and dashboards. My work connects the business question, the calculation, and the action a team can take, with particular experience in service operations and inventory reporting. SELECTED PROJECTS Finding an Ageing Support Backlog With SQL and Python Independent portfolio project | May - August 2026 Context: Built a reproducible support-operations analysis using a purpose-built synthetic dataset. The sample represents 600 unique tickets across eight weeks and three queues: onboarding, billing, and technical support. Business question: Which queue should a support lead investigate first when the open backlog is growing? My role: I designed the dataset and schema, wrote the SQL transformations, analysed the results in Python, and created a concise queue summary with recommendations. Method and decisions: - Started with 624 raw rows, including 24 duplicate ticket records, and retained 600 unique tickets after deduplication. - Checked ticket identifiers, timestamp order, queue labels, and the relationship between status and resolution dates. - Used a fixed reporting snapshot at 00:00 UTC on 29 June 2026 so the backlog calculation can be reproduced. - Kept open tickets in the backlog calculation and excluded them from resolved-ticket duration summaries. - Calculated ticket age and response intervals in elapsed hours, with the definition recorded alongside the queries. - Compared queue volumes with backlog age rather than treating the largest queue as automatically the most urgent. Sample findings: At the reporting snapshot, 96 tickets remained open. Of those, 58 were in the onboarding queue. The queue summary pairs that concentration with ticket-age bands so a lead can distinguish recent arrivals from older unresolved work. Recommendation: Prioritize an onboarding backlog review, assign an owner to the oldest unresolved tickets, and track incoming versus resolved volume before making a staffing decision. Portfolio materials: A synthetic CSV dataset, SQL queries, a Python notebook, a queue-comparison chart, and a metric glossary explaining the reporting snapshot and treatment of open tickets. What the project demonstrates: Data cleaning, reproducible reporting, metric definitions, exploratory analysis, and turning a finding into a focused operational question. Building One Weekly Inventory Report for Six Depots Data Analyst | October 2025 - February 2026 Context: Operations teams used three separate spreadsheets to report equipment availability across six depots. Different labels for available, reserved, and in-repair equipment made the totals difficult to compare. My role: I owned the data preparation, metric definitions, SQL model, Power BI report, and reconciliation checks. Depot managers agreed the operational definitions and tested the report against their local records. - Mapped source status labels to a shared set of equipment states. - Normalized depot identifiers and added checks for duplicate equipment records. - Defined available equipment separately from stock that was reserved or awaiting repair. - Created a Power BI view showing availability, exceptions, and the date of the last refresh. - Added a reconciliation page so managers could trace a total to its contributing records. - Documented the refresh schedule and assigned owners to recurring source-data issues. Outcome: The operations team adopted one weekly report in place of three independently maintained reporting spreadsheets. Managers gained a shared definition of availability and an exception list they could act on before the next reporting cycle. Portfolio presentation: A recreated schema, fictional example records, an annotated report layout, and an explanation of the reconciliation approach. EXPERIENCE Data Analyst Northbank Services | January 2023 - Present - Build SQL datasets and Power BI reports for service operations, inventory, and team planning. - Investigate differences between reported totals and source records, documenting causes and corrections. - Work with operational teams to define metrics, reporting cutoffs, and ownership of source fields. - Automate recurring preparation and validation steps with Python. - Present findings through concise summaries that connect the evidence to a decision or next investigation. Junior Data Analyst Northbank Services | August 2020 - December 2022 - Prepared weekly operational reports and reconciled spreadsheet inputs with database extracts. - Wrote SQL queries for recurring requests and documented calculation logic. - Built checks for missing values, duplicate records, and inconsistent categories. - Supported report users with explanations of definitions, filters, and source-data changes. EDUCATION BSc in Mathematics and Statistics University of Sheffield | September 2017 - June 2020 SKILLS Analysis: Business-question framing, exploratory analysis, data cleaning, metric design, reconciliation, descriptive statistics, and communicating findings. SQL: Joins, common table expressions, window functions, conditional aggregation, and date calculations. Python: pandas, Jupyter, matplotlib, repeatable data preparation, and validation checks. Reporting: Power BI, Power Query, DAX measures, Excel, dashboard design, and metric documentation. Working practices: Git, reproducible notebooks, stakeholder workshops, and clear handover notes. PROFESSIONAL FOCUS Operational analytics, service performance, data quality, and reporting that helps teams decide what to investigate and improve next.