Skip to main content

Four Database Integration Use Cases

Database integration is notoriously tricky to manage. If you’re avoiding it and running queries on your transactional databases, your databases are likely overburdened by workloads. If you’re trying to build your pipeline internally, your data engineers are likely spending countless hours writing scripts and juggling outages at the expense of fresh, accurate and reliable data. Regardess, transactional databases hold a lot of data that data teams can use to analyze and drive insights across the entire organization. If you’re not centralizing and analyzing your database data, you’re leaving valuable insights on the table.

Efficient, data-driven companies are leaving ETL (or ELT) to companies like Fivetran. As a result, they save time and stress, and can focus on adding value. As you’ll see below, regardless of industry, location or company size, businesses with reliable access to database data can derive insights and make informed decisions around revenue, customer journey, customer support, marketing channels and more.

1. Analyzing production data from multiple databases

Civil construction business Emery Sapp & Sons, generates a ton of data on a daily basis: the usual administrative, financial and labor workforce data, as well as data coming in from the field such as time spent on a job and a location, equipment runtime and utilization, and units of production – all stored within databases. When the business decided to migrate its data from PostgreSQL and SQL Server to a cloud data warehouse, the Director of Technology began by building scripts himself, but quickly learned that spending 20% of his time on pipeline management wasn’t sustainable or scalable.

With Fivetran, he offloaded the burden of data pipelines and built up-to-date dashboards for reports that used to take hours to create. Dashboards include fuel reports, revenue summary dashboards, accounts receivable dashboards, branch manager dashboards, and job summaries that show granular data points for individual jobs and enable the team to quickly identify large variances and react more quickly to potential problems. The construction industry operates on narrow margins, so using data to work faster and smarter to improve efficiency and gain a competitive edge has been key. Read the full case study.

2. Joining website data with transactional data to improve customer loyalty

Australia’s largest pet ecommerce store, Pet Circle, wished to move away from using database services on top of Google Cloud, and chose Fivetran to centralize its transactional data from MariaDB into its Google BigQuery data warehouse. By choosing Fivetran over building internally, Pet Circle saved months of engineering effort and ongoing maintenance that would have required an additional engineer.

By joining views of website sessions with paid data in a single transformed table, the business can understand how often customers visit the site, login, modify their subscriptions and more. This data helps Pet Circle improve loyalty and retention by adding customer value rather than transactional value. Read the full case study.

3. Understanding and predicting Customer Lifetime Value (LTV)

Branch Insurance, a home, renters and auto insurance company, brought on its data stack months before the company went live and needed a way to centralize transactional data from DynamoDB, the business’ most critical data source, into its Google BigQuery data warehouse to be able to produce reports within Looker. The co-founder and CTO found that Fivetran was the only solution to successfully replicate data from DynamoDB to BigQuery. Without Fivetran, the business estimates that it would have to hire a team of two plus people to manage data pipelines.

With its transactional data centralized, Branch can evaluate its marketing channels, understand customer acquisition costs and lifetime value, and modify campaigns and spend in real time. The company looks at metrics like 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 and chats, and more. Branch has also built models that estimate LTV based on the characteristics of purchases. Read the full case study.

4. Joining transactional and marketing data to allocate marketing budget effectively

Pre-owned vehicle ecommerce marketplace Cars24 used custom pipelines to bring data from its MySQL stores into a larger MySQL database, which functioned as the warehouse. The analytics team had to run its queries on top of MySQL, which caused the ‘warehouse’ to go down frequently. After estimating that it would take a team of five or six more than four months to build the initial pipeline, with considerable maintenance afterwards, Cars24 decided to trial Fivetran. Within two days, the MySQL database connector was up and running.

With Fivetran seamlessly piping data into Snowflake, crashing the database is no longer a problem. By joining transactional data with marketing data, Cars24 can understand channel performance and allocate budget in the most effective way. Read the full case study.

To learn more about our available database connectors and features, database integration best practices, and more, visit our Database Analytics Resource Center.

About Fivetran database connectors: Fivetran integrations create reliable pipelines from your transactional database into your analytics destination so that up-to-date data can be used to drive insights across all departments of your business. Our database connectors are easy to set up and have low-impact, micro-batched queries; log-based replication; automatic schema handling; and more. Learn more here.