Building a data foundation for a fast-growing fintech.

dopay provides digital payroll and prepaid card solutions to businesses across Egypt, giving workers access to financial services without requiring traditional bank accounts. Licensed by Egypt's Central Bank, dopay processes payroll for thousands of businesses across sectors including agriculture, construction, retail and healthcare.

As the business scaled, the volume and complexity of its operational data grew significantly. Leadership needed a way to bring that data together — securely and reliably — to inform decisions across the company, from commercial strategy to compliance and risk management.

---

The challenge

dopay's core application generates over 5 million new records a month across payroll transactions, card activity, KYC processes and customer onboarding. The data was locked inside operational systems, making it difficult for teams across the business to access the insight they needed.

The data itself carries significant sensitivity. It includes personally identifiable information (PII) for both companies and employees, alongside regulated financial data. Any analytics solution needed to handle this complexity without compromising on security or compliance.

Beyond access control, the raw data required substantial transformation before it could be useful. Financial transactions needed reconciliation and validation. Employee and company data needed to be aggregated into models that non-technical users could query without risk of exposing sensitive fields.

dopay needed a data platform that could handle high volumes, enforce strict access controls, and deliver reliable, queryable data models — all without slowing down the engineering team or creating operational risk.

---

The solution

We designed and built an ETL pipeline to extract data from dopay's application databases, transform it for analytical use, and load it into Google BigQuery as the central data warehouse. Looker was deployed as the BI layer, giving teams across the business self-service access to dashboards and reports.

Data extraction & recency

The extraction layer was built to handle the volume and variety of dopay's data sources, pulling in transactional, customer and operational data on a scheduled basis with built-in monitoring and alerting for pipeline failures. The pipeline runs every 15 minutes, giving teams across the business access to near real-time data — critical for a fintech where transaction volumes, card activity and compliance metrics need to reflect the current state of the business, not yesterday's.

Transformation & query performance

The transformation layer addressed the core complexity of the project. We built data models that aggregated raw transactional data into business-friendly structures — removing the need for analysts to write complex joins across normalised tables. Financial data was validated and reconciled as part of the pipeline, ensuring consistency between source systems and the warehouse.

The data models were designed so that the heavy computation happens during the transformation stage, not at query time. This means users can explore large datasets through Looker without triggering expensive queries, waiting for long-running reports, or needing to understand the underlying table structures. It keeps compute costs predictable and, crucially, removes the friction that typically deters operational users from engaging with data tools at all.

Security & access control

Security was embedded throughout. We implemented role-based access controls within BigQuery and Looker to restrict who could see what. PII fields were redacted or masked depending on the user's role, ensuring that teams could access the data they needed without exposure to sensitive information they shouldn't see. All access to the data environment was secured and auditable.

Visualisation & self-service BI

For visualisation, we deployed Looker as the BI layer, building dashboards and reports that gave teams across the business self-service access to key metrics — from transaction volumes and revenue to customer onboarding rates and compliance KPIs.

---

The results

dopay now has a production data platform processing over 5 million new records a month, with data flowing from operational systems into BigQuery and surfaced through Looker dashboards across the business.

Teams that previously relied on ad-hoc database queries or manual spreadsheet exports now have self-service access to validated, up-to-date data models. Commercial, operations and compliance teams can answer their own questions without engineering involvement.

The security model ensures that sensitive financial and personal data is handled appropriately — with field-level redaction, role-based access, and a full audit trail of who accessed what and when.

The platform has become a foundational piece of infrastructure for dopay, supporting everything from board reporting to regulatory compliance and day-to-day operational decision-making.