When it comes to tech, many ideas gain popularity quickly as a halfway solution to a problem with longstanding challenges.
One of the latest for business intelligence is data blending. Data blending is a trend that has been going on for years. But now new tools have added capabilities meant to save analysis time to digest and sort data.
When you blend data, you combine data from multiple database sources into a single table or worksheet of a given solution.
It's a relatively quick and straightforward way to extract value from multiple data sources, meant for discovering patterns between data types without the time and expense of addressing traditional data warehouse processes.
Certain big data use cases demonstrate where data blending makes a lot of sense. Aaron Cohen, General Manager of Strategic Planning and Analytics at Audi, for example, explained how the auto maker leverages data blending.
Audi combines data from external influences on car purchases, including social and political, with customer influences trends, such as tastes and lifestyle.
From those results, it can help form strategies and actions to keep Audi competitive with other premium automotive brands.
As Cohen’s example suggests, data blending enables uncommon types of data to coexist in a model. This is one reason why big data blending has attracted interest among businesses.
Discovering ways to bring together unusual data feeds in one repository facilitates new data combinations that influence strategic decisions.
So one of the primary tasks of data analysts is to determine the "blends" of the data that must be useful. There are three aspects that an analyst must consider for establishing a data blending table:
- Determining the number of data connection to other databases
- Identifying the data sources within the solution, which can mean understanding the data types (CSV, SQL databases, etc.)
- Assessing the number of records being managed
Data blending is sometimes confused with data joining, but there is a significant difference.
Joining involves combining data from the same source. Blending involves combining data from different sources, such as sales data from Excel and demographic data extracted from a Google Analytics profile.
There are times when joining may be more beneficial – when the similar source data is easier to pull.
But if there are clear complications from the volume of records presented, blending may be a better choice. The key is evaluating how the data is used and frequently updated.
There are a number of tools that have worked to blend data sources and offer a semblance of a shared data repository in the process.
Tableau, for example, offers the capability to automatically blend two data sources when the software detects a common field among two or more data sources.
Another data blending solution is MicroStrategy Analytics Enterprise suite. The company has been launching refinements to its blending capability to combine data from multiple sources without requiring a separate integration tool or advanced technical assistance.
The latest enhanced data blending and on-the-fly features offer faster ease of use.
Other popular tools include, Pentaho, which added “at the source” data blending capabilities back in 2013, and Alteryx, also a popular choice among firms that regularly leverage business intelligence dashboards and models.
Data blending capability isn’t a permanent fix for all big data challenges. Not all types of data are amenable to being blended with others.
But as the number and type of data sources continues to grow, so too will the desire of business users to select data blending in assessing various data correlations in models and building dashboards based from the data.
Title image by torbakhopper.