Global Software Company Centralizes & Refreshes Data from M&A in Snowflake
Customer's Challenge
A leader in the CRM software industry recently acquired a complementary business and desired to centralize all business data into a single source of truth using Snowflake. The newly-acquired business’s data was housed in a separate Snowflake instance that was scheduled to be decommissioned soon so the customer needed support to accurately & efficiently migrate the data by the deadline.
phData's Solution
phData proposed an accelerated 12-week engagement to assist the customer in moving the newly acquired business’s data into the customer’s existing Snowflake instance. Once approved, phData deployed a night & day cycle of migration that included the use of the Data Source tool from the phData Toolkit to ensure data freshness and pipeline stability.
Results
phData successfully moved the newly acquired business’ data into the client’s Snowflake instance by the deadline. The customer now has a streamlined tech stack and a single source of truth housed within Snowflake that they can lean on to make data-driven business decisions.
Additionally, phData improved the client’s data freshness and data pipeline stability, which will aid in data accuracy and reporting for their many business units in the future.
The Full Story
A global CRM software company had recently acquired a new business and desired to centralize all business data within its own Snowflake instance. The acquired business had been operating under its own Snowflake instance that would soon be decommissioned, risking data loss if it was not quickly migrated elsewhere.
The customer reached out to phData in hopes of partnering on a speedy data migration project. They had previously worked with phData on other data and ML projects and knew phData had a team of experts who could streamline and accelerate the work.
Operation Move to Snowflake in 12 Weeks
The teams joined forces and began the project in late October. Their goal was to have all of the business data (including that of the newly acquired business) centralized in their Snowflake instance by the end of January. This was no small task, as the customer had 300+ dbt models, 200 tests, and several terabytes of data to be moved. Moreover, there’s often additional leg work involved in “lift and shift” engagements, like ensuring source data exists and addressing schema drift.
phData began by meeting with the many stakeholders of the customer. The data that needed to be centralized was relied upon by more than 20 of the customer’s business units, making it a highly visible project. This required phData to work closely with the customer to prioritize the data sets and pipelines to be migrated and keep a continuous flow of communication to a multitude of stakeholders through the engagement.
To meet the tight but crucial 12-week deadline, phData utilized a day & night cycle of work that included both their US and India-based data teams. The US team utilized the phData Data Source tool to validate data freshness and pipeline stability, and the data team in India kept the replication process going after hours. Thanks to this approach, the teams were able to complete 30-40 models per week!
With phData’s help, the customer was on track to meet their deadline and produce improved data outcomes for their business units. Many of the business units had previously relied on manual reporting processes, and phData was able to refresh data sources and stabilize pipelines that would automate many of these processes.
Take the next step
with phData.
Learn how phData can help solve your most challenging data analytics and machine learning problems.
