How a Semantic Layer Improves Tableau Performance for Dimensional Analysis

How a Semantic Layer Improves Tableau Performance for Dimensional Analysis

I have always been a fan of Tableau, and we have written before on the powerful combination of Tableau and AtScale. With Tableau expanding on  its BI and Analytics position with their acquisition, the potential for building powerful BI programs on AtScale and Tableau are all the more powerful.  

Tableau entered the market eighteen years ago differentiating itself from legacy BI platforms with its versatility. The platform supported the blending of data from different sources to produce dashboards and reports that supported organizations in making smarter, data-driven decisions, making it a widely embraced platform across organizations and  industries.  

With Tableau, we’ve seen just how wide those uses go. Retail organizations use the platform to analyze sales and financial metrics, manufacturing teams analyze their inventory and logistics, and software organizations use it to analyze customer churn and key growth KPIs.’s Tableau for CRM offering embeds powerful visualizations, augmented by Einstein analytics, directly into the user experience of SFDC.  

Below, let’s examine a few of the ways Tableau and AtScale can help organizations make data-driven decisions.

Dimensional Analysis in Tableau

Dimensional analysis, a term that is commonly associated with legacy BI and the world of OLAP, is really a universal analysis concept that continues to be extremely important to the world of enterprise analytics. 

At AtScale, we frequently support organizations that deploy Tableau alongside Microsoft Excel. Excel Pivot Tables are a powerful dimensional analysis tool that is undisputedly the most widely used analysis tool for ad hoc data analysis. Our customers will frequently use Tableau for dashboarding and executive reporting with Excel the workhorse for analysts. 

The rise of Microsoft PowerBI as a dimensional analysis oriented analytics platform has challenged Tableau’s position as the leader in business intelligence.  In addition to its deep integration with the Microsoft stack, Power BI gives business users an interactive analytics experience within a dimensional analysis framework that ensures consistency and supports analytics governance – both core to the concept of “Self Service BI.” 

The AtScale Semantic Layer provides a powerful dimensional analysis platform that tools like PowerBI, Excel, and Tableau can sit on top of, allowing organizations to better align data consumption across different use cases. This approach ensures governance, consistency, and performance, thereby supporting self-service BI for data-driven organizations. 

AtScale also leverages this dimensional understanding to build and manage aggregates pulled in previous queries to accelerate response for Tableau users. This approach avoids the need to scan an entire warehouse for every query supporting “speed of thought” dimensional analysis within Tableau

Growth in Cloud Data Platforms

With enterprise data continuing to grow exponentially, cloud data platforms (either cloud data warehouses like Snowflake or data lakes like Data Bricks) are becoming the standard for both large and small organizations. After all, cloud data platforms can easily store the petabytes of data modern enterprises can accumulate. Therefore, it is highly preferable to connect BI directly to cloud data platforms to both avoid the work and maintenance of ingesting data locally while also ensuring a holistic approach to organization information . 

While Tableau has integrations with all major cloud platforms, the challenge comes with delivering an interactive BI dashboard on massive cloud-scale data sets. If Tableau has to scan billions of rows of raw data, a fairly straightforward query can take hours or even days to execute. This problem is exacerbated when many users are attempting to query the same dataset simultaneously — an issue known as “query concurrency.”

As discussed above, by deploying AtScale between Tableau and cloud data platforms, business users can execute high performance queries directly on large data sets. By removing the primary causes of slow time to insight (raw table scans and query concurrency), a semantic layer can  reduce query response time from days to minutes, enabling businesses to make data-driven decisions in real time.

This exponential improvement in speed can make all the difference, allowing businesses to reach fresher insights and respond in kind. 

The Challenge of Dimensional Analysis on Cloud Scale Data

Building robust dimensional analysis capability on cloud scale data is complex on any BI platform. Without applying the governance and consistency of formal dimensional analysis, pulling data from different sources can result in confusion. For example, if it’s unclear as to which column from which table in a data store actually represents the value that a user is trying to use in a report, then there can be no clarity as to how to apply that data for the organization. 

By deploying AtScale between Tableau and a data warehouse, users are insulated from the complexity of cloud scale data. The business oriented data model presented to Tableau users can be easily understood and queried with confidence (i.e. the returned data will reliably match that queried from other data platforms.) 

This single source of truth (SSOT) is vital for building a data-culture and is also helpful in coaching a broader set of users to become proficient with using Tableau to answer their questions. By building out a clear and transparent layer, everyone in the organization can work with clarity to draw the same conclusions from the same information. 

Using Tableau and AtScale to Accelerate Data-Driven Decisions 

AtScale and Tableau are a proven combination for enabling powerful dimensional analysis capabilities on cloud-scale data. Combining the flexibility and elegant user interface of Tableau with the benefits of a robust semantic layer, organizations can empower themselves with self-service BI and the growth of a data-culture.  

To learn more about how AtScale supports dimensional analysis in Tableau, you can download this datasheet. And to learn more about how used AtScale and Tableau to build an analytics program on Google Big Query, you can explore this case study.


GigaOm Sonar Chart - semantic layers and metrics stores

AtScale Developer Community Edition