Can too much self-service be too much of a good thing? Opening up access to data will yield diminishing returns if users are still in the dark about its provenance. Alation, which offers cataloging and collaboration tools that help users discover and share insights about the data they are using, has added a starter edition tailored to Tableau users that will alert them to issues that could impact whether and how they use that data. It is a useful step for ensuring that Tableau users who are enjoying their freedom with self-service don't cause collateral damage in the process.
What’s behind the data you’re using?
When it pioneered the modern wave of self-service business intelligence (BI) visualization, Tableau struck a nerve: business end users were hungering for the means to take control of their data and analytics without having to rely on IT. But while it is one thing to provide a tool, the challenge is to ensure that the business outcome of providing self-service is advantageous to the business in the long run. The challenge of successful BI self-service is ultimately tied to the age-old, and still valid, IT maxim: garbage in, garbage out.
Alation is part of a wave of tools that are harnessing machine learning to help demystify BI data. For Alation, it is harnessing crowdsourcing to help end users understand the provenance of the data they are using, and how to query it. While Alation has typically been deployed with Hadoop, its foothold with Tableau customers made the company aware of the need for users within that environment to gain more visibility into the data they were working with.
That resulted in a new, starter edition product that goes against Tableau data sources as opposed to Hadoop. The new offering, Alation Tableau Edition, integrates directly into the Tableau user interface (populating tabs and projects with pertinent information) to address one aspect of data governance: ascertaining the provenance of the data that they are using.
Here’s how it works. For instance, issues in data lineage, such as an incomplete ETL data transformation run, can be highlighted in red; if the issue is sufficiently urgent, an email to the appropriate recipient(s) can be generated. The product also provides capabilities for data stewards, who can promote specific workbooks to Tableau users or move data sources between different Tableau projects. The latter capability can be especially useful when new projects enter the test phase and need data sources to populate the shakedown runs.
Alation’s move comes as other providers of tooling for curating big data (e.g. data preparation) are also looking to address points of pain in more established installed bases of BI users. For Alation, the introduction of a tool that can help tag or highlight data sets – good or bad – in Tableau is a useful step toward ensuring that newly self-reliant BI visualization end users don't wind up digging themselves into holes caused by reliance on questionable data.
On the Radar: Alation harnesses crowdsourcing and machine learning to speed data access, IT0014-003097 (January 2016)
2017 Trends to Watch: Big Data, IT0014-003164 (November 2016)
A Practitioner's Guide to Self-Service BI and Analytics, IT0014-002967 (December 2014)
Tony Baer, Principal Analyst, Information Management