Branch Centralizes Transactional Data for Reporting

Key Takeaway

After testing many pipeline solutions, Branch Insurance found Fivetran was the only product to successfully centralize its DynamoDB data. To build, maintain and scale without Fivetran, Branch would need to hire at least two additional employees. Branch’s Looker dashboards display near-real-time information for acquisition cost and lifetime value, enabling the business to modify campaigns and spend in real time.

Data Stack

• Pipeline: Fivetran
• Connectors: AWS Lambda, DynamoDB, Google Analytics, Google Sheets, SendGrid, SFTP, Stripe
• Destination: Google BigQuery
• Business Intelligence Tool: Looker

Branch Insurance is a home, renters and auto insurance company. By filling out their name and address, clients can bundle home and auto insurance in seconds, completely online. The business launched statewide in Ohio in August 2019, is live in five states, and plans to be national by 2021.

Launching a Business With the Right Architecture

Branch brought on its data stack months before the company went live. After building the ETL tooling for multiple businesses in the past, Joe Emison, co-founder and CTO of Branch, wanted to have the right data architecture in place from the start:

Building backoffice solutions as a startup is asking for technical debt that will cost you in the future as you grow. I learned that the largest difficulty in building your own pipelines is keeping them running overtime as schemas change and things break. When I set up Fivetran and saw it run, I thought: "Wow, this company is special – they get it."

The most critical data source for Branch is DynamoDB, which houses its transactional data. The business needed to bring the data into Google BigQuery to be able to produce reports within Looker. Emison tried many products, and landed on Fivetran:

The short story is that there is only one product that successfully gets data from DynamoDB to BigQuery and that’s Fivetran. I tested many other products and it wasn’t even close.

A Data Pipeline That Just Works

Emison estimates that it would have taken him a full work week to build the synchronization logic for the DynamoDB connector. Without the connector from Fivetran, he would have to hire a team of two plus people to manage data pipelines as the business grows: “We’ve basically saved a whole team of people by not babysitting our data pipelines. I want to ensure that my employees are spending 90% of their time on projects that improve our client’s experience.”

In addition to time and resource savings, other benefits of Fivetran include:

• Minimal configuration
• A limited choice model that adapts to schema changes and runs smoothly
• Connectors that just work
• Functions connectors to connect to other APIs

Reporting on Critical Data

The most important information that Branch collects is around marketing channel performance. “We test a lot of different marketing channels and we need to understand our customer acquisition costs and lifetime value,” Emison explains. To get an understanding of acquisition cost, Branch looks at metrics that join data from multiple systems, including:

• Cost to get prospects through the door
• Cost of the data the business has to pull to determine the insurance price
• Sales resources spent on calls, chats, etc.
• Marketing spend info synced via a Google Sheet

Branch has models that estimate LTV based on the characteristics of purchases. A dashboard displays near-real-time data for acquisition cost and lifetime value, enabling the business to modify campaigns and spend in real time, which Emison recognizes as critical: “If we’re unhappy with the LTV or acquisition costs from a specific channel, we can respond to that information quickly – and that’s magic.”

Want to join marketing and transactional data to improve your acquisition and lifetime value analytics? Trial Fivetran for free today.  

About Fivetran: Shaped by the real-world needs of data analysts, Fivetran technology is the smartest, fastest way to replicate your applications, databases, events and files into a high-performance cloud warehouse. Fivetran connectors deploy in minutes, require zero maintenance, and automatically adjust to source changes — so your data team can stop worrying about engineering and focus on driving insights.

About BigQuery: BigQuery is Google’s serverless, highly scalable, enterprise data warehouse designed to make all your data analysts productive at an unmatched price performance.

About Looker: Looker is a modern platform for data that offers data analytics and business insights to every department at scale, and easily integrates into applications to deliver data directly into the decision-making process.