Job Description

We are hiring a Data Analytics Engineer to own a fast-moving insight platform being built on Microsoft Fabric for one of our clients, a business process outsourcing (BPO) company based in Cape Town. This is a full-time, on-site role at the client's Cape Town offices.

The client has a stable operational system of record that handles billing, payroll, and monthly KPI reviews. That system stays as it is. What they need is an insight platform on top of it: daily operational dashboards, fine-grained splits across channels and services, custom metrics, and the ability to ship dashboard changes in days rather than weeks. This role owns that platform end-to-end.

To be clear about what the role is and isn't:

• You own the daily insight platform, including ticket, call, and chat events, channel splits, custom metrics, ad-hoc analysis, and dashboard iteration.
• You do not own billing or payroll calculations. Those continue to live in the existing system of record.
• You build the reconciliation between the two systems so they agree on shared aggregates, with the system of record being right by definition.
• You are the person managers come to when they want a new metric, a new split, or a new dashboard view, and you ship those changes in days.

This is a hands-on technical role with full ownership and direct access to leadership. There is no committee between you and shipping a change.

Responsibilities:

• Run the morning data health check and confirm overnight ingestions succeeded across multiple campaigns and dozens of daily inputs.
• Investigate and resolve broken ingestions: changed sheet templates, late files, schema drift, validation failures.
• Onboard new campaign data sources, typically a SharePoint folder, a daily template, or an exported report.
• Write and maintain dbt models that transform raw data into trusted KPI tables.
• Build and refine Power BI dashboards for internal departments and, later, client-facing reports.
• Review the reconciliation dashboard and investigate any aggregates where the insight platform and the system of record disagree.
• Field requests from operations, QA, workforce management, finance, and campaign managers for new metrics, dashboard changes, and ad-hoc analysis.
• Keep the KPI dictionary current: when a definition changes or a new metric is added, it gets documented and modelled.
• Manage the Power Automate flows that validate incoming files and alert when something is wrong.
• Own data quality. When a number on a dashboard looks wrong, you are the person who finds out why.
• Manage Azure costs, Fabric capacity scaling, and licence allocations.
• Document everything so the platform doesn't depend on knowledge in your head.

Qualifications:

• Strong SQL, comfortable writing and optimising complex queries with joins, window functions, and CTEs.
• Hands-on data warehouse experience with Microsoft Fabric, Azure Synapse, Snowflake, BigQuery, or Redshift.
• Power BI fluency: DAX, data modelling, row-level security, and performance tuning.
• Experience designing data pipelines that pull from APIs, files, and databases.
• Comfort with Git-based workflows (commits, branches, pull requests).
• Demonstrable experience owning data quality: not just building, but keeping things accurate over time.
• Strong written communication. You'll be documenting heavily and explaining things to non-technical stakeholders.
• Comfortable being the only data person in the room and making decisions accordingly.

Nice to Have:

• dbt experience (Core or Cloud). If you know SQL well but haven't used dbt, we'll give you ramp-up time.
• Microsoft Fabric experience specifically (still rare, and Synapse experience is fine).
• Power Automate or Logic Apps for orchestration.
• Experience building insight platforms alongside a separate system of record without disturbing it.
• Experience migrating off MS Access. This is a real migration, not a greenfield build.
• BPO, contact centre, or customer service operations background, so metrics like AHT, FRT, CSAT, and QA scoring feel intuitive.
• Experience with Zendesk, Intercom, Klaus, or similar helpdesk reporting.
• Python or PowerShell scripting for occasional automation.
• Microsoft Purview or other data governance tooling.

Technical Stack:

• Warehouse and lakehouse: Microsoft Fabric (OneLake, Lakehouse, Warehouse, Data Pipelines)
• Storage: Azure Data Lake Storage Gen2
• Ingestion: Azure Data Factory, Power Automate, SharePoint, Logic Apps
• Transformation: dbt Core (SQL-based modelling, version controlled in Git)
• Visualisation: Power BI (datasets, reports, row-level security, paginated reports), Zoho Analytics
• Source systems: Zendesk, Intercom, Klaus, Aircall, HubSpot, Zoho, and other helpdesk, telephony, and CRM platforms, plus client Google Sheets
• Governance and security: Entra ID, Azure Key Vault, Microsoft Purview (planned)
• Code and docs: Git, GitHub, Markdown, SharePoint wiki

Benefits

We offer competitive compensation, a learning and certification budget, and a supportive, low-ego team that values work–life balance and growth. You'll get full ownership of the platform, a direct line to leadership, and a Microsoft 365 with Copilot licence plus full Fabric and Azure access. This role is on-site in Cape Town.