GCP Big Query by Bravas Technology
Scalable SafetyCulture analytics built natively in BigQuery.
Features
Structured Warehouse Data Layers
SafetyCulture API data is organised into Bronze (raw), Silver (curated), and Gold (analytics-ready) datasets within BigQuery. This layered approach separates ingestion from curated analytics outputs and ensures reliable data promotion between processing stages. It aligns with Medallion architecture principles used across enterprise analytics platforms.
BigQuery SQL Transformations
Transformations are implemented using BigQuery SQL to flatten JSON structures, apply schema mapping, and create relational datasets. Incremental merge logic ensures inspections, actions, users, and activity records are updated efficiently. This creates structured fact and dimension tables ready for analytics.
Incremental Batch Processing
Watermark ingestion ensures only new or changed records are processed during each execution cycle. Batch identifiers enable monitoring, reconciliation, and data lineage tracking across processing layers. This maintains reliable and efficient enterprise refresh cycles.
BI Visualisations & Dashboards
Bravas designs and builds dashboards in Looker Studio or other BigQuery-connected BI tools. KPI definitions, semantic reporting models, and operational dashboards are implemented on top of curated Gold tables. These dashboards provide visibility into inspections, actions, compliance, and operational performance.
Description
Bravas delivers a scalable, production-ready integration from SafetyCulture into BigQuery.
SafetyCulture data is extracted via REST API, landed into Bronze storage within Google Cloud, transformed into curated Silver datasets, and modelled into structured Gold warehouse tables inside BigQuery.
The solution supports:
• Full and incremental endpoint ingestion
• Watermark-based updates using SQL merge/upsert patterns
• Batch-level traceability across Bronze, Silver, and Gold layers
• Structured SQL transformations orchestrated with scheduled queries (GoogleSQL DDL/DML supported)
• Stored procedures for reusable transformation logic, control-flow, and KPI calculations
• Partitioning and clustering optimisation to improve performance and reduce scanned data
• Relational modelling for inspections, actions, users, groups, sites, and activity logs
• Business-ready marts for analytics and BI
• BI dashboards designed and built by Bravas in Looker Studio or other BigQuery-connected BI tools
• Watermark-based updates using SQL merge/upsert patterns
• Batch-level traceability across Bronze, Silver, and Gold layers
• Structured SQL transformations orchestrated with scheduled queries (GoogleSQL DDL/DML supported)
• Stored procedures for reusable transformation logic, control-flow, and KPI calculations
• Partitioning and clustering optimisation to improve performance and reduce scanned data
• Relational modelling for inspections, actions, users, groups, sites, and activity logs
• Business-ready marts for analytics and BI
• BI dashboards designed and built by Bravas in Looker Studio or other BigQuery-connected BI tools
BigQuery serves as the final warehouse layer, optimised for scalable, governed, and high-performance enterprise reporting.
Testimonials
Tyler Mason
CTO at El Jannah
The Bravas team delivered an excellent end-to-end data solution for El Jannah, integrating SafetyCulture, Employment Hero, Deputy, and Sonder data into our Azure environment. Their structured approach and technical expertise gave us a reliable, scalable data foundation for reporting and operational insights. The implementation was professional, well-managed, and aligned to our broader data strategy.
Pricing
* The pricing here is for display purposes only. You should contact the partner for the most up to date and correct pricing information. We do not take any responsibility for this pricing information, which is provided by our partners. Pricing last updated: 12 Mar, 2026 12:00AM
FAQ
BigQuery is required for datasets and tables, and a durable Bronze landing zone is typically implemented using Google Cloud Storage before batch loading into BigQuery via load jobs. If streaming-style freshness is required, the Storage Write API can be used (directly or via supported connectors).
Transformations are primarily implemented in GoogleSQL using scheduled queries for orchestration and repeatable execution (including DDL/DML). Incremental updates use MERGE for upserts, and reusable KPI/business-rule logic is implemented using stored procedures and procedural scripting constructs.
Dataform is optional where you want version-controlled, dependency-managed SQL workflows for BigQuery transformations. Dataflow is optional where managed batch/stream processing is required, and it supports writing to BigQuery using Storage Write API modes (including exactly-once semantics in supported configurations).
BigQuery pricing is typically a combination of storage plus query/compute costs, with compute available as on-demand (per data processed) or capacity-based models using slot reservations for more predictable performance and budgeting. Cost control is supported by best practices such as partitioning, clustering, and governance around query patterns.
Yes—Bravas designs and builds BI dashboards on curated Gold datasets in Looker Studio and other BigQuery-connected BI tools. Looker Studio supports connecting directly to BigQuery tables, views, and custom queries, and BI Engine can be used to accelerate interactive dashboards where required.
