[00:00:00]
Introduction
Armando Lopez opens the Unlock the Value webinar series by highlighting the urgency of digital transformation. In today’s fast-moving world, ensuring that technology investments drive measurable business impact is critical. This session will explore how to streamline operations and modernize infrastructure with real-world strategies from Improving’s consultants.
He introduces Don Sawyer, Director of Data Engineering at Improving Minnesota. Don will walk through a multi-ERP consolidation case study involving Snowflake and SAP, showcasing Improving’s collaborative and technical capabilities in solving complex enterprise challenges.
[00:01:02]
Understanding The ERP Challenge
Don begins by framing the central issue: companies often operate with multiple ERP platforms, each with different structures, purposes, and levels of customization. These fragmented systems introduce operational inefficiencies and limit business insight.
Common ERP Scenarios:
Legacy Systems – Old platforms deeply embedded in core operations but hard to retire.
Customized ERPs – SaaS or licensed platforms heavily tailored to specific processes.
Specialized ERPs – Tools used for narrow functions, like finance or acquired business units.
Resulting Challenges:
Data duplication and inconsistent master records
Misaligned reporting and limited visibility
Delays in analytics due to manual data stitching
Bottlenecks in business operations, especially in seasonal industries like agriculture
Technical and organizational silos that block innovation
Don shares how these barriers led their client, an agribusiness organization, to seek a unified, scalable, and future-ready ERP platform centered on SAP.
[00:06:04]
Building the Unified Solution
Improving partnered with the client’s internal teams—not just delivering a solution, but co-developing it—to consolidate ERP data into SAP Central Finance. This unified platform would allow the client to modernize financial operations while enabling future transformation efforts.
Solution Architecture (High-Level):
Source ERPs – A mix of batch and CDC (Change Data Capture) systems.
Snowflake – Centralized cloud data warehouse used for processing, staging, and modeling data.
DBT & Streams/Tasks – Used to curate raw and enriched datasets.
Hightouch (Reverse ETL) – Sent clean, standardized data into SAP for real-time access.
Key Concepts Introduced:
Data Contracts – Each ERP team aligned their output to a consistent schema, making it easier to integrate.
Parallel Development – Multiple teams worked independently using shared rules and standards.
Staged Views – Allowed teams to isolate changes and prevent downstream disruption.
This architecture enabled the organization to combine near real-time and batch data from different ERPs into a single finance function within SAP.
[00:20:19]
ELT Design & Data Migration
Don dives deeper into the Snowflake-based ELT architecture and how it enabled scalable data processing and historical migration.
Three-Tier ELT Approach:
Raw Layer – Ingests ERP data using batch or CDC methods.
Process Layer – Enriches the data with metadata (timestamps, deletes, etc.).
Staging → Model Layer – Applies business rules per ERP, standardizing outputs into SAP-ready formats.
Reverse ETL with Hightouch: Hightouch triggered updates individually for each ERP stream, enabling differentiated refresh cadences (e.g., every 5 mins for one ERP, daily for another).
Historical Migration (Cutover Process):
Loaded balance snapshots into SAP at a defined cutover point.
Views were parameterized to limit data visibility until go-live.
Allowed teams to stage data early while ensuring clean reporting post-cutover.
This modular design allowed the client to preserve historical integrity and transition smoothly without service disruption.
[00:27:42]
Collaboration & Expertise
Don emphasizes that this wasn’t just a technical lift—it required strong cross-team collaboration and domain-specific problem-solving.
Improving’s Key Strengths:
Data Management Expertise: Deep knowledge of DBT, SQL, testing, and Snowflake design.
Agile Development: Used CI/CD pipelines and modular coding to isolate changes and minimize regression risks.
Standardization: Shared templates and peer review processes ensured consistency across ERP streams.
Observability: Built dashboards and alerting tools (e.g., for out-of-balance documents) using Power BI and Snowflake views.
Cross-Team Collaboration:
Business Partners: Co-created and validated complex business rules (e.g., intercompany flows, multi-currency logic).
ERP SMEs: Helped decipher nuanced platform behaviors and validate assumptions.
Scrum Leads: Reworked story structures and sprint planning for better visibility and forecasting.
By blending consulting excellence with hands-on development, Improving built both the solution and the trust that underpinned it.
[00:42:46]
Outcomes & Closing
The project resulted in significant operational improvements and laid a foundation for future innovation.
Business Outcomes:
Faster, more consistent financial reporting
Unified ERP data inside SAP with access to advanced analytics and automation features
Reduced manual processes and reliance on niche ERP experts
Increased business agility for growth and seasonal demand shifts
Final Takeaways:
Strong partnerships between consultants, developers, and business users drive better outcomes.
Modular, standards-driven architecture simplifies maintenance and onboarding.
Investing in platform thinking and unified data flows accelerates digital transformation.
Armando closes by thanking Don and previewing the next session in the series, focused on Databricks and Confluent for a modern data platform.
