Our Client
BDUK is part of the Department for Science, Innovation and Technology (DSIT) and is responsible for the rollout of gigabit-capable broadband and the expansion of 4G mobile coverage in hard-to-reach areas of the U.K.
Business Challenges
The U.K. government body responsible for improving the nation’s digital connectivity, Building Digital UK (BDUK), needed faster, more accurate, and higher-quality data to improve a range of business processes. These included geospatial data and analysis, more accurate and timely subsidy allocation and supplier payments, and clearer programme monitoring.
BDUK is part of the Department for Science, Innovation and Technology (DSIT) and is responsible for the rollout of gigabit-capable broadband and the expansion of 4G mobile coverage in hard-to-reach areas of the U.K.
The new data capability was required to identify target premises that were underserved by the existing broadband market. BDUK could then facilitate access to gigabit-capable broadband for homes and businesses that were not otherwise in market plans. BDUK’s unusual position as both a funder and delivery enabler, as well as a coordinator between government, suppliers, and communities, demanded robust digital systems and data management.
Unconnected systems
BDUK had been held back by unconnected systems and inconsistent data that required heavy manual intervention to put right. Overly bespoke data pipelines and heavy spreadsheet manaagement meant there was no single source of data truth due to an absence of shared data definitions and data models.
This situation was compounded by legacy systems, including SQL-based and manual Excel processes, that couldn’t scale with the growth of BDUK’s programmes to deliver real-time assurances.
Project Gigabit
BDUK therefore consulted Google Cloud, who recommended creating a cloud-based semantic layer with a single source of truth, based on their industry-leading flagship data product, BigQuery, to provide data-as-a-service for all stakeholders. Project Gigabit was underway, so BDUK had to deliver in tandem with programme delivery.
To be a success, this would mean:
Replacing fragmented manual payment assurance with automated, standardised processes.
Ensuring scalability for growing volumes of supplier payments.
Improving visibility, governance, and auditability.
Replacing manual, error-prone pipelines with automated ingestion and validation.
Delivering data in accessible formats so geospatial data analysts had more time to solve high-value problems.
Ensuring data governance and compliance with cyber/InfoSec standards.
Providing real-time transparency into data quality and reporting
Google and SoftServe together
SoftServe placed “Agile delivery” at the project foundations, starting with mapping the “asis” process (from audit findings) and incrementally automating key assurance steps. There was then a phased rollout across key data and payment validation pipelines.
By embedding assurance in systemised workflows, payments would become more efficient. It would enable faster and more accurate decisions about relevant properties, stronger programme governance, and improved supplier confidence.
Modern platform
Under BDUK’s strategic direction, SoftServe placed “Agile delivery” at the project foundations, starting with mapping the existing systems and processes and incrementally automating key assurance steps. There was then a phased rollout across key data and payment validation pipelines.
The first step was the development of Data Backbone, a modern, cloud-native ingestion and validation platform created by SoftServe & BDUK. This would enable GIS Payments Automation, using Data Backbone as the core assurance tool.
Next, SoftServe extended Data Backbone pipelines to ingest, validate, and reconcile payment data and then automate further key programme data assurance checks to reduce manual effort. Dashboards were built for real-time monitoring of assurance process status.
This meant SoftServe built a single source of truth across all BDUK functions, which enabled business users to directly query the data through the new semantic layer. It laid the groundwork for BDUK to embark on building AI tools and agents, and it automated hundreds of data quality and assurance checks at the point of submission.
Standardised workflows
The introduction of standardised workflows provided efficiencies across finance and commercial teams and replaced manual reviews with structured, validated data. It also improved visibility with end-to-end dashboards and clearer ownership.
Together, this produced a substantial reduction in manual effort that allowed staff focus to shift from manual data validation to higher-value activities, such as property identification and treating exceptions, rather than repetitive checks. Pipeline acceleration significantly reduced ingestion times.
The technology used to bring this about included:
Modular design
The modular design ensures reusability for future programmes, all because of strong tri-partner collaboration between BDUK, SoftServe, and Google Cloud.
It now supports transparency in one of the U.K.’s largest public digital infrastructure programmes, while Data Backbone, initially built for supplier data ingestion, proved flexible enough to directly address the need for greater internal data consistency.
The success of the Data Backbone project demonstrated how a modular, cloud-native architecture can pivot to not only meet new assurance challengesn but also provide security for the sensitive data that was being processed.
Better governance
Access to faster, more reliable data meant better governance of the multi-billion-pound programme and reduced operational risk. BDUK has been able to more quickly scope and define its interventions and has increased supplier trust due to reduced risk of errors and delays in supplier payments.
The outcome also reduced the risk of overpayments or delays in subsidy allocation and was delivered on time and on budget. More importantly, by making the Data Backbone open source (available on request) for all Public Sector entities, it will enable savings for any future projects with other departments.
SoftServe was invited to the BDUK project as it is a Premier Google Cloud Services Partner with a track record of proven results. This status recognises our deep and recognized expertise in enabling clients to develop cloud-native applications through leveraging advanced technologies like AI/ML and data analytics. With over 700 certified Google Cloud professionals, we’ve successfully delivered hundreds of Google Cloud projects ranging from Fortune 100 enterprises to digital-native startups.
