Skip to content

Data Analytics for the Information Prosumer

by - 1. April 2018

Encounters with Maestro and Wrangler

A few days ago, I came across the short preview of an analytics software from Tableau that is called Maestro. This software – which is being called project by Tableau as it is supposed to become part of the tableau software products – is focussing on preparing data before the analyst is analyzing the data. Preparing data starts with importing data from one or more data sources and combining them via different set operations. Traditionally this is done via textual commands (e.g. written in SQL, MDX), whereas in Maestro it is done via a graphical interface which is visualising data sources, the ways they are combined and the transformations the user is applying to the data. This interface is presented in a pane which is called the „Flow“ and it reminds me of the visualisation of data operations in the Trifacta Wrangler which is focussing completely on the data preparation phase for data analysts.

There are even more parallels between Project Maestro and Trifacta. A large part of work before an analyst starts the analysis itself, he needs to invest into exploring the technical nature of his data and clean them up. Real world data are messy: columns having the wrong data type, data are missing, same things written in different ways, and so on. Everyone working with data knows that much time is invested into this kind of work. Smart analytics software like Maestro and Wrangler are supporting the user in multiple ways. Some examples are assistive technologies like:

  • automatically analysing imported data and categorizing, structuring it and displaying the data type
  • profiling the data values and showing the distribution of current values in a histogramm, so that the user can discover missing, mismatching or inconsistent values more quickly.

I have discussed some of those features in analytics software in more depth in the article „Von der Tabellenkalkulation zur assistenzgestützten visuellen Analyse“ for the book „Qualität und Data Science in der Marktforschung“. In my article I identify important changes in interface paradigms of Business Intelligence software from an Interaction and UX Designers point of view.

From Tabular Calculation to Assistent-supported Visual Analytics

Written in summer 2017, the article shows and explains that interfaces not only of BI software but of data analytics software in general are changing towards:

  • deep and more intelligent usage of visualisation of data in all phases of the analysis
  • integration of assistive technologies and libraries using artificial intelligence and machine learning
  • Interface design making use of state-of-the-art visual and interaction design principles

Both Trifacta Wrangler and Project Maestro are part of an overall change in the way, user interfaces for analytics software are designed. And these changes are connected with changes in the type of user who is using these kind of tools. Previously the user types Data Analyst and Technical Architect have been the almost exclusive user of Analytic tools and handed over the results to the information consumer.

Nowadays, the information consumer has evolved into the new user type of information prosumer, as I call him along the lines of Tofflers „prosumer“ definition. An information prosumer is a person who is analysing data on his own, navigating through all the available data, exploring the data ad-hoc, and adapting information depth and width of his dashboard to his current and individual needs, without having been trained in expert tools for data retrival, ETL, BI, architecture of data warehouses and the like. Keywords connected with this already emerged type of user are „democratisation of data“ and „citizen data scientist“.

In my above mentioned article I discuss in depth why this user type matters a lot to software engineers, UX designers and product owners and what is making the difference to the information consumer which will still exist but did not have much relevance in professional analytic tools.

One important thing: all of the book is written in German only … sorry! If you know German and are willing to pay for my article, here are the data:

Title: Von der Tabellenkalkulation zur assistenzgestützten visuellen Analyse
Author: Lothar B. Blum
— Link for purchase e-paper will be made availabe in May 2018 —

The article will appear in this book:


Title: Qualität und Data Science in der Marktforschung
Subtitle: Prozesse, Daten und Modelle der Zukunft
Editors: Bernhard Keller, Hans-Werner Klein, Thomas Wirth
Publisher: Springer Gabler

The complete book will be published end of May 2018. You may (pre-)order here:



From → Allgemein

Schreibe einen Kommentar

Kommentar verfassen

Trage deine Daten unten ein oder klicke ein Icon um dich einzuloggen:

Du kommentierst mit Deinem Abmelden /  Ändern )

Google+ Foto

Du kommentierst mit Deinem Google+-Konto. Abmelden /  Ändern )


Du kommentierst mit Deinem Twitter-Konto. Abmelden /  Ändern )


Du kommentierst mit Deinem Facebook-Konto. Abmelden /  Ändern )


Verbinde mit %s

%d Bloggern gefällt das: