Table of contents
About the client
Based across two continents, the client is an expert in the monitoring and management of large-scale telecommunications networks.
They provide a comprehensive range of services that allow network operators to run, optimise and manage networks on a 24/7/365 basis.
Business challenge
For many years, a Senior Technical Director prepared the monthly SLA report covering up to 10 locations. While leading the process, he also relied on ad-hoc support from other team members who provided inputs from different systems.
Data came from several sources (including ticketing, operations, communication tools, Excel and SharePoint) making the process time-consuming and complex.
Preparation of each report took up to five working days every month. Due to its complexity, the report was usually finalised 20 days after month-end.
As the company continued to grow and prioritise timely decision-making, the need for a faster, more scalable reporting process became clear.
Fabric-first reporting platform
As the client’s trusted technology partner, we designed and delivered a new end-to-end SLA reporting in Microsoft Fabric, with OneLake serving as the single source of truth across the entire data landscape.
The solution spans the full data journey: from acquisition and processing to the delivery of SLA KPIs, providing a unified, scalable foundation for analytics and decision-making.
Our collaboration in numbers
90%
time reduction in SLA reporting
450 h/year
saved annually by automating manual tasks
1 day
time-to-decision on SLA insights immediately after month-end (vs. 20 days before)
0.25 FTE
recovered from manual work for higher-value management tasks
Up to 10
service locations with unified SLA reporting
Architecture and data flows
With our solution, data from various source systems and file formats is automatically ingested using Fabric Dataflows Gen2, Pipelines, Lakehouse, SQL and Python-based Notebooks. Scheduling and orchestration ensure data is collected reliably each month.
A layered (medallion architecture) processing model transforms raw data step by step: from bronze (raw), through silver (cleansed), to gold (KPI-ready). Shared dictionaries and calendars standardise definitions across all reported locations.
To secure trustworthy results, we additionally designed data quality rules in a Quality Monitoring Module: a central mechanism that applies rules for data completeness, consistency, and duplication. Any issues trigger alerts and quality scores, allowing the team to spot anomalies early and ensure that SLA reports are both reliable and trusted in decision-making.
Reporting & presentation layer
For fast, interactive performance with minimal latency, we built a lightweight reporting layer in Power BI, connected directly to OneLake using Direct Lake mode. The solution includes:
- SLA Executive – a high-level view of key performance indicators (KPIs), with 12-month trends, benchmarking across cities, and dynamic filtering.
- Assets Acceptance – an operational dashboard supporting assets handovers, aligned with SLA compliance metrics.
- Service Reliability Dashboard – an operational dashboard that supports outage management and allows automatic monitoring of central and distributed infrastructure availability, also aligned with SLA compliance metrics.
- Quality Monitoring Module – adds rules to the data quality dashboard for validations, scoring, and detailed data with links to the source system.
Main benefits of our partnership
- 90% reduction in reporting effort, saving approximately 450 hours per year.
- Reduced time-to-decision from up to 20 days to same-day (Day 0) data availability.
- Improved data quality and transparency through harmonised SLA definitions, automated checks, and repeatable processing.
- Recovered approximately 0.25 of a full-time equivalent (FTE), allowing a senior director to focus on higher-value strategic work.
- Enabled new analytical capabilities, including 12-month trend analysis and city-to-city performance benchmarking.
Technologies used in the project
