Success story

LearnVest: Building an end-to-end analytics platform

By Kit Germeroth

LearnVest, now run by its parent company Northwestern Mutual, is a financial planning company that helps people manage and navigate finances. The organization’s mission is to educate and empower individuals to manage their money to achieve specific goals, and they have a number of tools and apps that help users track their spending and financial goals.

An End-to-End Analytics Platform

LearnVest needed to modernize and streamline their entire analytics program. Prior to working with Keyrus, LearnVest had several databases with different types of information, making it difficult to locate specific data, understand sources, and monitor quality. The project also required a heavy degree of enterprise security from a tech perspective. Because LearnVest handles sensitive personal data (referred to as PII) for its customers, their data must be protected and encrypted in certain ways to protect these customers.

The goal was to create an integrated system between all of the company’s data sources in order to track the customer journey from awareness to conversion and product use.

Choosing the Best Technology

Our team integrated with LearnVest’s data engineering team and set out to build a new end-to-end analytics platform. The project began with an architecture design phase in which we mapped out technology options and tools with special emphasis on which modeling technique to use for the data warehouse. Our team implemented a data vault instead of the traditional warehouse and created proof of concepts (POCs) for each tool or technology. We then compiled a list of requirements with weighted priorities for each to determine which technology was best.

Building a Data Vault

The next project phase involved a longer POC in which we implemented all of the selected tools. We collected data from three sources into the data vault to assess the performance of the chosen tools. After positive results, we moved forward with the POC and began integrating more data sources for the longest phase of the project.

Our team implemented Tableau, Enterprise DB, and Snap Logic. LearnVest uses AWS as its cloud platform, and also has Splunk, Alation, and Spark Stream deployed in their organization.

Supporting the Data Science Teams

Prior to the engagement, LearnVest’s data science teams didn’t have easy access to the data. We worked with their teams to build a customized solution for each team that took data from the vault. With business-specific data marts, each team could use data science to answer their own questions in ways they couldn’t before. Since they no longer have to worry about data cleansing, they can focus on actually getting insights from the data.

Results

The end result is a robust system that integrates all data sources for fast and easy access. The Planning Experience team uses the system to recommend financial products and plans to users based off answers from a self-service tool. As the core of LearnVest’s services, this greatly helps them deliver a better, more personalized customer experience.