Case Study

Mortgage Lender Accesses Customer Data in Near-Real-Time With Cloud-Based Information Architecture

The Customer’s Challenge

A major midwest mortgage lender—funding $60B+ loans each year with more than 10,000 employees—needed to convert leads more effectively using the data they already had. Their legacy infrastructure consisted of 6 disparate sources which didn’t allow real-time access. This disconnected data meant team members were working with limited insights as they talked to leads. The solution: operationalize their current marketing and research data to better convert leads into revenue.

phData’s Solution

Our expert cloud team stepped in to get the client running on Snowflake by creating an information architecture designed for the cloud while constructing exhaustive data pipelines. This project aims to support multiple dynamic customer facing teams and their changing data needs. Therefore, phData’s approach to automating data pipelines following CI/CD best practices equipped the client with what they’ll need to add new data sources with a templated architecture ensuring no lag in the business value output when scaling. We created a scalable and cloud native unified analytics platform allowing for real-time composite customer lead score generation.

The Full Story

As a major mortgage lender, the client places customer service front and center in its operations. But after nearly three decades of service, the company was pulling customer data from six disparate sources.

With thousands of loan originators on the team and hundreds of branches across the nation, our client needed a cloud data platform to build out the lead scoring that the company envisioned. 

They quickly landed on Snowflake with the platform’s ability to handle complex ingestion quickly, securely and effectively. 

Why phData?

As an Elite Snowflake Partner, phData was at the top of the list as the Snowflake account team made recommendations for an implementation partner. With our data engineering experience, the client knew they’d be in good hands as they moved data to the cloud for the first time. 

Disparate data—and a clear case for centralization

When they decided to move to Snowflake’s cloud data platform, the client had six separate data sources that they wanted to collect in one place. On top of internal systems and data sources, these included:

The client’s primary goal was crystal clear from the start: convert leads more effectively using the data they already had. The solution (from the expert view of the phData data engineering team) was equally clear: create an information architecture designed for the cloud. But it’s not as simple as just “moving” everything to Snowflake. 

An Information Architecture Designed for Long-Term Success

As the data engineering team prepared to move the client into Snowflake, they kept three goals in mind:

After standing up the Snowflake environment for the client, phData’s data engineers moved into setting the client’s customer-facing and dev teams up for success in the long term. 

Using Snowflake, Azure Data Factory, and Azure DevOps, the team created an ingest architecture to connect external systems with both Snowflake and data consumers (like PowerBI, the data visualization software used by the client). 

To ensure everything ran smoothly at handoff, our data engineering team also set up all relevant integrations and developed CI/CD for automated testing for the data pipelines, making ‘maintenance’ much more straightforward for the client after handoff. 


Previously, the client’s infrastructure would delay data access by a day or more. Now, phData’s client can access data in 4-hour increments, with the opportunity to move to 15-minute increments in the future. 

We can check the “mission accomplished” box on the original goal: easy to access, centralized customer data. But our data engineering team went above and beyond to also ensure the client would have what they need as they continue to use cloud data in more complex ways. First, the team transformed the data in Snowflake to meet the enterprise standards of their information architecture. Now, the client can easily add in new data sources using the templated architecture and CI/CD— while their admin team can maintain and manage each pipeline based on the customized run book we delivered at the end of our engagement.

With a sound data pipeline and a fully implemented Snowflake environment, the client will be able to engage customers in new ways from now and into the future. 

Take the next step
with phData.

Looking to move to Snowflake or get more use out of your customer data? Learn how phData can help solve your most challenging problems. 

Data Coach is our premium analytics training program with one-on-one coaching from renowned experts.

Accelerate and automate your data projects with the phData Toolkit