Expert Interview (Part 2): Decideo’s Nieuwbourg Talks Trends in Big Data, IoT & Data Visualization
In Part 1 of our two-part conversation, Phillipe Nieuwbourg (@nieuwbourg) talked about the founding of Decideo, interesting developments in data, what tools and skills you need, and mistakes to avoid to get the most out of data. In today’s part 2, he addresses what businesses should be doing with data today, and down the road, and trends in Big Data, IoT and data visualization.
In general, what should businesses be doing from a technology standpoint today to prepare for the ways they’ll use data down the road?
First: collect and store data. Without fresh and accurate data in your data lake, you can’t analyze anything. It’s the first step of a data culture. Of course, privacy and regulation are key to collecting the right data you will be able to analyze.
From a technology standpoint, it means to create what we call a “data lake” – a place where data will be stored, ready to be used by analytics applications we already know or for future needs. And of course, I repeat it, to verify and protect data from their sources to the data lake, to be sure we will be able to use it to generate value.
What are the most promising new IoT applications you’re observing in business today?
IoT is fantastic. Humans can generate data, but they have limits that sensors don’t have. Objects, sensors, can generate thousands of data every minute. It’s an inexhaustible data source.
Predictive maintenance, consumer behavior analytics, video recognition for security applications … there’s no industry that will avoid the IoT wave. If you don’t know yet how IoT will transform your industry, focus on it! It will, you are just late!
Related: The IoT-Big Data Convergence
Why is effective data visualization so critical to an organization’s ability to understand their data? What should organizations look for in high-quality data visualization tools?
Imagine you have analyzed terabytes of data. You found something, a behavior, a trend, a pattern… how will you bring this to your management? Will you, like a research scientist, produce a black and white text presentation with complex equations? Certainly not the best way to convince your general manager. Just one image is better than a report or a long text explanation. But how to choose the right graphical visualization? How to create the Wow effect in your boss’ eyes? That’s where data visualization software come.
Do you know that Excel can generate only two types of graphic? There’s plenty of highly understandable graphics that your management will love, but that you can’t do in Excel. Just remember how Charles Joseph Minard created in 1869 the famous Napoleon’s march to Moscow graphic. The data visualization you will choose doesn’t depend on your data, it depends on your message. And you need an advanced software to do it.
Related: Visualization for Big Data
How has the way companies are able to visualize their data evolved in recent years? What developments interest you in the world of data visualization right now?
We’ve moved from data visualization to data storytelling.
Like in Hans Rosling famous TED presentations, you must tell a story. People never remember the statistics you put in your PowerPoint, they will remember the story you told them. And to transform your boring slides into a stunning story, you will need to apply storytelling techniques to data analysis. Animated data, storytelling techniques, you have the keys – the same used in Hollywood movies, or to write the scenario of House of Cards next season. By the way, do you know that Netflix is the first data driven movie/series producer?
What trends or innovations in the field of Big Data, data visualization and IoT are you following right now? Why do they excite you?
I continue to follow the data storytelling field. Actual software is a first generation. I miss a lot of functionalities. There’s actually no real data storytelling software on the market. I hope to discover one very soon.
And I’m really interested in deep learning, specially applied to IoT video feeds and photos analysis. If we can automatically “understand” a situation, we can automatically suggest an action. It can be a recommendation, a human intervention, but if we help people to understand data more quickly, we will make a step forward to the “augmented intelligence” I was talking before.
What’s one piece of advice you find yourself repeating to organizations over and over related to Big Data? One takeaway you think every organization should hear?
It’s not a technology project! Big Data is not a technology issue, it’s all about business. Stop buying technology before having understood and measured your data value. The result of Big Data analysis can be very small. Imagine a model that every morning gives you the list of the 10 prospects you must call today because they are ready to sign. Perhaps it’s Big Data analysis, but the result is a list of 10 names … small data. Don’t focus on technology, focus on business.
And don’t tell me Big Data is expensive. It’s never too costly. Because you will always prototype, often with open source software or with tools you already have, and anticipate your ROI before investing in high level infrastructure. Big Data will never be a charge, it’s will always be a source of revenue. If you accept to move step by step and focus on business challenges. And I wish the best of luck to all our readers. Data Driven Economy is a fantastic opportunity for the actual and next generations to generate value, and do things better.
To learn how to put your legacy data to work for you, and plan and launch successful data lake projects with tips and tricks from industry experts, download Syncsort’s Building a Data Lake Checklist Report!