Centralizing SaaS Data into a Unified Analytics Platform
Synthflow.ai, a SaaS company in the AI automation space, works with multiple data sources to power its business. Data was distributed across platforms, including Stripe, HubSpot, GA4, Google Ads, Linear, Pylon, and MongoDB. With each source operating independently, Synthflow lacked a centralized system to integrate this data and transform it into business-ready models.
The leadership team needed dashboards that could serve as a single source of truth, ensuring reliable and consistent insights for customer support, sales, marketing, finance and product teams. To achieve this, Synthflow partnered with Datum Labs to design and deliver a modern data stack built for automation, scalability, and cross-team usability.
Outcomes That Matter
- Unified Data: Different SaaS and product data sources centralized into BigQuery.
- Reliable Reporting: Consistent KPIs across sales, marketing, finance, product and support.
- Automated Pipelines: Fivetran ingestion removed manual data handling.
- Business-Ready Models: dbt layered transformations ensured accuracy and governance.
- Cross-Team Adoption: All customer-facing and leadership teams transitioned to the new dashboards.
- Scalable Architecture: A foundation ready for future growth and new integrations.
Why Change Was Essential for Synthflow.ai?
Scattered Data Sources
With information spread across Stripe, HubSpot, GA4, Google Ads, Linear, Pylon and MongoDB, teams lacked a unified system to measure performance. Each department relied on its own reports, making it difficult to align insights across the business.
Inconsistent Metrics Across Teams
Because data was siloed, sales, marketing, finance and product often relied on different definitions of the same KPIs. This created reporting misalignment and made leadership decisions slower and less reliable.
Manual Reporting Burden
Analysts spent significant time exporting data from SaaS platforms, cleaning it in spreadsheets and assembling performance reports. This slowed down reporting cycles and increased the risk of human error.
Limited Real-Time Visibility
Customer-facing teams needed near real-time dashboards to track adoption, resolve tickets and monitor trial-to-paid conversions. The previous setup could not refresh fast enough to support these critical decisions.
Scaling and Governance Risks
As Synthflow’s customer base expanded, so did its data. Without a scalable warehouse and structured transformations, performance bottlenecks and governance challenges became inevitable.
The Modern Data Platform We Delivered
Datum Labs built a future-ready analytics platform with scalability, automation, and usability at its core.
1. Automated Data Ingestion with Fivetran
We connected Fivetran to all major SaaS and database sources, including Stripe, HubSpot, GA4, Google Ads, Linear, Pylon and MongoDB. These automated pipelines ensured that fresh data was continuously loaded into BigQuery without engineering overhead.
2. Structured Modeling in dbt
We implemented dbt to transform raw data into business-ready models. Transformations were organized into staging, intermediate and mart layers, grouped by domain (sales, marketing, product, finance, support). This layered approach ensured consistency, modularity and reusability. Business logic was embedded in models, aligning KPIs across all teams.
3. Orchestration and CI/CD with Dagster
Using Dagster, we automated workflows, monitoring, and scheduling. Git-based deployments allowed for version control, testing, and CI/CD of dbt models, ensuring smooth rollouts from staging to production without manual intervention.
4. Interactive Dashboards in Metabase
Finally, we delivered Metabase dashboards customized for each business unit:
- POC Dashboard: Measured customer adoption and engagement during proof-of-concept trials.
- Enterprise Dashboard: Tracked enterprise customer activity across agents, calls, and workspaces.
- Customer Support Dashboard: Monitored agent performance and resolution trends.
- Marketing Attribution Dashboard: Linked HubSpot, GA4, and Google Ads to attribute leads and trials to campaigns.
- Freshworks Dashboard: Reported sub-account activity across agents, customers, and calls.
These curated dashboards gave every team a single source of truth, eliminating the misalignment that plagued the old system.
The Stack and Strategy Behind the Success
- Data Infrastructure & Warehousing: BigQuery (scalable, cloud-native warehouse).
- Data Ingestion & Integration: Fivetran (automated pipelines from SaaS and databases).
- ETL & Transformation: dbt (structured modeling and business logic).
- Orchestration: Dagster (workflow management, monitoring, scheduling).
- BI & Visualization: Metabase (interactive dashboards and reporting).
The Transformation Journey
Synthflow.ai’s transition to a centralized, automated data platform transformed the way the company measures, analyzes, and acts on data. By consolidating multiple SaaS sources into BigQuery, applying structured transformations with dbt and delivering user-friendly dashboards in Metabase, every department gained access to reliable, consistent insights.
The impact has been significant:
- Marketing can now attribute trials to campaigns with confidence.
- Sales and finance rely on a single revenue source of truth from Stripe and HubSpot.
- Support tracks performance in real time and reduces ticket backlogs.
- Executives monitor adoption, engagement and revenue growth in a single view.
With Datum Labs’ expertise, Synthflow.ai now operates on a next-level data foundation that scales as it grows. Data silos are gone, reporting is automated and every team works from the same trusted metrics.
Is your business ready to unify your data into a single source of truth? Let’s make it happen.
.svg)
%20(1).webp)
