Companies from every industry vertical, including finance, retail, logistics, and others, all share a common horizontal analytics challenge: How do they best understand the market for their products? Solving this problem requires companies to conduct a detailed marketing, sales, and finance analysis to understand their place within the larger market. These analyses are designed to unlock insights in a company’s data that can help businesses run more efficiently. They also share a common set of tasks: Collecting a variety of data sources, integrating those into a centralized data platform, and developing analytic capabilities that support the development of reports and dashboards.

The most common solutions to these challenges require a large suite of tools that each are used to perform a single step of the process, then pass the data along to the next tool. This requires data engineering teams to learn, build, operate, and monitor a data pipeline with a large number of points of potential failure and significantly lengthens the development process. Many customers use Fivetran’s automated data integration to solve this challenge—it can reduce complexity by automating many of these processes and help your teams reduce the time to value by enabling earlier analysis. This post shares an example of a customer who is doing exactly that, as well as step-by-step instructions on how to implement something similar in your environment.

Connecting the dots across systems

Let’s consider the predictive sales analytics use case of Brandwatch, who wanted to focus on understanding the ties between customer events on their application and their Salesforce data. This type of example has implications that are broadly applicable across other industries and solutions, where the requirements are still focused on centralizing data into a common data platform across a variety of sources such as a customer relationship management (CRM) platform, event data, and other marketing data sources you may have.

In the case of Brandwatch, understanding the ties between their tracked events collected through Mixpanel, a service that helps companies understand how users interact with their products, and how those features impacted sales and accounts required centralizing that information into a single, central data warehouse. The most pressing question from Brandwatch’s product team was simply whether new features were getting adopted. After these initial findings, more options become available to answer tangential questions, such as refining standalone product features to improve customer retention, seeing how previous customer feedback tied into new features, and the result of implementing new features in the sales pipeline.

#google cloud platform #data analytics

How Brandwatch automated its data pipelines with BigQuery and Fivetran
1.50 GEEK