The adoption of the cloud introduces a profound shift in how we think about data: it is no longer a static asset stored in a predictable, self-contained environment. Instead, data is in a constant state of motion, driven by the evolving needs of the business and the ever-expanding capabilities of cloud platforms. This perpetual movement is a core theme of modern data architecture, creating a set of common scenarios that every organization must navigate.
This constant flow—to, from, within, and between clouds—means that moving data is not a one-time event, but a continuous operational reality. The decision of where your data should live is intertwined with the question of what you need to do with it. Understanding the drivers, strategies, and architectural patterns for these movements is fundamental to building a resilient and agile modern enterprise. The following story of two political organizations on the eve of an election provides a stark, real-world illustration of what happens when the need to move and process data collides with the limitations of an aging architecture.
<aside>
A Tale of Two Data Infrastructures—The 2016 U.S. Election
A stark real-world example of cloud migration drivers, technical debt, and architectural readiness can be found by comparing the data operations of the two major U.S. political parties leading into the 2016 presidential election.
"It's not the system's fault it wasn't working… It wasn't built to last a long time or have the number of users it ended up having.”
— Robby Mook, Clinton campaign manager, on the DNC’s aging infrastructure
The Democratic Party: A System on the Brink
By 2016, the Democratic National Committee’s (DNC) central data platform — a monolithic, on-premises data warehouse running on Vertica (HP’s columnar database) — was failing. During peak campaign moments, the system would crash for hours or even days at a time, stalling field operations and forcing staff into ad-hoc workarounds. Campaign staff reported sleeping in their cars while waiting for essential voter-targeting queries to complete.
This failure represented a textbook case of technical debt — and of suffering from early success. What began as a pioneering advantage in the 2008 and 2012 cycles — rooted in groundbreaking platforms like the on-premises Vertica data warehouse and Narwhal, a unified data integration system that stitched together voter, volunteer, and donor records into a single actionable profile, enabling unprecedented targeting precision — became a liability by 2016.
The core issue was a dramatic shift in the scale of use. In 2008, the system was primarily the domain of specialized data scientists and analytics teams. By 2016, it had become a self-service platform for a far broader spectrum of campaign staff across field, communications, and digital teams. This democratization of access accelerated local decision-making but multiplied demand beyond what the aging infrastructure could sustain.
The very architecture that had once delivered a decisive edge now buckled under its own weight. Under pressure, the DNC undertook a risky, mid-cycle migration to a modular, cloud-native architecture on Google BigQuery — a disruptive move that diverted resources at the height of the campaign.
The Republican Party: A Calculated Investment
In contrast, the Republican National Committee (RNC) had spent years building a more durable data infrastructure, having learned from the Democrats' decisive data advantages in the 2008 and 2012 elections. In 2010, the RNC partnered with the firm Data Trust to consolidate its voter files into a scalable and tested system.
By 2016, the Trump campaign inherited this mature, vendor-supported infrastructure. While the campaign famously contracted Cambridge Analytica for polling analysis and ad visualization, the RNC's data repository remained the core voter file of record. This stable foundation allowed the campaign to effectively run massive A/B-tested ad campaigns and perform granular voter targeting, while their Democratic counterparts were preoccupied with firefighting and rebuilding their core systems.
The Outcome: An Infrastructure Deficit
Hillary Clinton later described the DNC’s data operation as “bankrupt,” stating that upon securing the nomination, she had “nothing to build on.” According to former Obama innovation chief Michael Slaby, the problem wasn’t just complacency — it was a failure to modernize over the years. He remarked: “Technology doesn’t sit still for 10 years.” The RNC, motivated by its “Growth & Opportunity Project” report after the 2012 election, had made a sustained investment in its data capabilities.
This case illustrates that architectural readiness is not about a single tool but about long-term strategic discipline — sustained investment in scalable systems, vendor relationships, and a culture of maintenance and modernization. Early success can create powerful momentum, but without continuous evolution, it can also harden into the very technical debt that undermines future performance.
</aside>
The story of the 2016 election illustrates the most urgent driver for cloud migration: the time-eroded platform — an infrastructure that was once cutting-edge but, through years of under-investment, can no longer meet operational needs, all while competitors evolve rapidly. The DNC’s on-premises system reached this point by 2016; it had to migrate or face total operational failure. While such urgent cases are powerful catalysts, most organizations move to the cloud for a broader set of strategic reasons, including:
Once the decision to move is made, the next question is how. The "Six Rs" is a widely adopted framework that outlines the common strategies for migrating applications to the cloud.