BizCover Google Analytics Ingestion and Modelling

In mid-2023, Google officially decomissioned their Google Analytics Universal Analytics (UA) product and reverted all their customers to the new Google Analytics 4 (GA4) product. Although there was plenty of notice given for this change, the migration path for customers required a rebuild of their Google Analytics setup in order to be GA4 compliant, as well as training up the business to understand the new metrics that were created in GA4. A good example of the changes was the introduction of the new ‘Engagement’ metric in GA4 to more accurately track how engaged a user is to the content of your website.

Our client, like many others caught on to these changes quickly and proactively built out their GA4 capabilities ahead of the hard stop date set by Google. They noticed however that their downstream BI processes for reporting were going to be heavily impacted by this change, especially because some of their core metric calculations for customer website interactions were affected. This is where they sought out Altis to assist them with sourcing GA4 data for their data warehouse and to re-build the datasets required for their Google Analytics reports used daily by the business.

The Vision

Altis was engaged by the Australian Maritime Safety Authority (AMSA) to design and deliver a modern cloud-based reporting and analytics environment (RAE).

The RAE was envisaged to provide a centralised single source of truth for reporting and analytics, drawing data from AMSA’s diverse operational systems’ data holdings. Within the RAE, data from sources would be standardised, integrated, and transformed to support a suite of data products. These data products would then be able to be shared across business areas, driving better-informed risk-based decision making and more efficient operations. The RAE would additionally be capable of reliably surfacing data into business intelligence and advanced analytics tools for reporting and data science.

The problem

Google Analytics is one of the core data for BizCover’s data platform, playing a pivotal role in daily operations and the strategic allocation of marketing resources to drive leads and conversions. A considerable challenge is that BizCover operates not one site but a dozen sites which requires GA4 conversion at the same time.

The challenge arose when BizCover faced a mandatory conversion to the newer GA4 from UA with a limited time window for all sites. This transition was critical not only for maintaining operational efficiency but also for ensuring marketing is targeting the right cohorts for the continued generation of business pipelines. In essence, it directly influenced the company’s bottom line and day-to-day operation.

The opportunity

BizCover saw the mandatory change as an opportunity. Two immediate opportunities were:

• To refactor existing solutions using new GA4 granular data to improve accuracy.
• To build new models using enhanced granularity data by GA4 as a new foundation to unlock addition web traffic insights.

The solution

The creative and practical solution came through a collaborative effort between Altis and BizCover’s team after detailed planning of parallel site refactoring and testing strategy. Given the time constraint, the focus was on devising a solution that minimised breaking changes to existing solutions while providing additional data models to harness the added capabilities of GA4 for future insights.

The solution involved bringing in a dozen sites from Google Analytics with the usage of BigQuery via Fivetran to Snowflake. Once data is landed in Snowflake, standardisation to all ingested sites is applied. Transformation logic, creation of new models and integration with other BizCover’s data are performed with dbt. End-user analytic reporting is via Power BI.

The outcome

The above practical solution was able to be delivered in a short amount of time and approved for production just days before the hard stop date for UA. The GA4 solution was created parallelly while UA data were still ingested. This has allowed the business to compare and test legacy UA solution against new GA4 solution for a decent amount of period which has provided confidence in the new GA4 data.

The impact is that there was no downtime for day-to-day decision-making and in some instances more accurate results. Two more accurate results are noticed in the success of UI UX funnel & experiments and marketing attribution. This results in an improved allocation of resources on sites as well as marketing spending to drive more traffic and interest from potential customers.

Tech stack

• Fivetran
• Google BigQuery
• Snowflake
• dbt
• GitHub
• Power BI

Ready to modernise your analytics stack and unlock deeper insights from your data? Altis Consulting helps organisations harness platforms like Google Analytics 4, Snowflake and Power BI to create trusted, insight-driven reporting solutions. Contact us today to find out how we can help you streamline your data pipelines and accelerate business performance.