Companies who want the flexibility of changing their data will pay for it later when the data is incorrect or unexpectedly changes but they cannot trace why.

In this sense, Excel’s strength is also its weakness. To begin with, anyone with write-access to an Excel file can edit it, potentially causing a data governance nightmare. But this approach fails to scale for several reasons. If there are any data issues, Excel is relatively easy for analysts to troubleshoot on their own. Excel is ubiquitous among analysts and easy to understand, making it a natural fit as a preferred data source in the beginning. Most early adopters of Tableau use Excel as a data source and host those files on a platform like SharePoint. Red Flag Practice #2: Files Hosted on SharePoint In this case, better to move that workload to a back-end database. There are many good uses for custom SQL, but this is not one of them.

But when you are abusing custom SQL by doing extensive data integration in that step, the dashboard will be impossibly slow. In many cases, the performance difference may not be noticeable. With a custom SQL connection, Tableau treats the entire connection as a nested SQL query, wrapping all the filtering, grouping, aggregations, etc. When you use the Data Source pane as intended, Tableau understands the structure of the data and constructs efficient queries based off this understanding. Why should custom SQL be avoided unless necessary? Well, Tableau’s data engine is very smart. Another example of just because you can do something, doesn’t mean you should. This can put a tremendous strain on dashboard performance since this is not what Tableau was designed to do.

The opportunity for this madness presents itself when the organization tries to use Tableau as a data integration tool to both transform the data and combine multiple sources together.

You read that right: 50 pages! It didn’t take long for us to identify why the dashboard stalled out. Once, we were brought in to help performance-tune an impossibly slow dashboard only to find that another consulting firm had written a 50-page SQL script inserted into the custom SQL. Over the years, we have seen this repeatedly abused by novice Tableau developers who come from a data engineering or SQL analyst background. When you go the Data Source pane of Tableau, there is a button titled New Custom SQL that allows you to write SQL for your data connection instead of clicking and dragging the tables and views for your joins. If any of the following apply to you, let’s talk! Red Flag Practice #1: Custom SQL I’m going to list five red flags when it comes to Tableau practices. Once an organization becomes big enough, it is likely time to start investing in a modern cloud data platform, so Tableau dashboards can be fed with trusted and governed data from a reliable source, such as a data warehouse or data mart. But some requests from Tableau users instinctively set off a red flag, signaling we should probably discuss how the data is getting into Tableau and whether the approach is a scalable, long-term best practice. Some organizations are more mature than others when it comes to data-not at all a problem. As any organization grows, it is imperative that it take a hard look at how data flows, what rules govern the exchange of data, how data gets transformed and who has access to the data:Ībove: Is your data trusted and governed?Īs consultants, we provide guidance and solutions for clients no matter where they are along their data journey. At the very top of the blueprint diagram is the overarching principle that data should be trusted and governed. Last year, Tableau came out with Tableau Blueprint, a methodology for building the capabilities of a successful, data-driven organization. So many options! However, just because you can use Tableau a certain way doesn’t mean you should.Ībove: So many options, but not all data sources are equal! The Tableau Blueprint By my last count, Tableau has over 80 native connectors to a range of sources. In part, it has accomplished this by providing its user base with a tremendous amount of flexibility when it comes to data sources. Tableau is the #1 visual analytics platform in the world.

Tableau prep best practices