Disruptive Analytics: Charting Your Strategy for Next-Generation Business Analytics

But it's not good news for everyone. Despite this tidal wave of data being produced by the digital economy, the leading organizations within the data analytics industry are struggling. Companies like SAP, IBM, Oracle, and Microsoft are laying off staff and seeing their sales either decline or stagnate.

Innovative new business models are disrupting the world of analytics, shaking up the more established companies by reinventing the ways that data gets used.

This book aims to help companies both big and small avoid being sunk by this disruption. "Disruptive Analytics" argues that businesses will struggle and go into decline unless they seize the opportunity to profit from the innovations that are disrupting business analytics.

The book challenges us to redesign our business models and processes, to realign our people and update our technology so we can benefit from these disruptive innovations. To help us do so, it surveys seven areas of disruption in detail, and shows us how to turn each threat into an opportunity.

So who would be interested in this book? You might think that a book about disruptive analytics would have a narrow audience of CIOs, data analysts, and IT consultants. And for sure, people in these roles will want to read this book. But think again about how data now touches all our lives and roles. Department heads, product developers, strategic decision makers, and even HR professionals will all find useful information in this book.

So, whether you manage a multinational or drive a taxi, the chances are that you work with data, business intelligence, or analytics at some level. And if you do, this book has relevance for you.

Thomas W. Dinsmore has more than three decades of experience with advanced analytics and machine learning. He's a leading authority on this topic and has worked as an analytics consultant for more than 500 clients around the world. He's co-authored two other books – "Modern Analytics Methodologies" and "Advanced Analytics Methodologies" – and he also publishes The Big Analytics Blog.

So, keep listening to find out how the open source business model works, how disruptive the cloud has been for leading analytics providers, and how to profit from innovation and disruption.

"Disruptive Analytics" comes in at 262 pages, and it's divided into 10 chapters, with a short introduction.

The book gets underway by presenting seven technology trends that are seriously upending the analytics world, disrupting every link in the business analytics supply chain, and transforming how businesses gain competitive advantage through data. These technologies are the subjects of the seven main chapters of the book.

The first is simply called "Fundamentals," but while it might be about the basics, be warned. Right from the get-go, the information here is dense and technical, and there's a lot to take in.

In this chapter, the author considers the conundrum that lies at the heart of this book – why established and respected technology companies are faltering when analytics as an industry is growing.

Take IBM, for example. IBM is recognized as an industry leader in data quality tools, data warehousing and advanced analytics, and yet in 2016 the company reported its 16th consecutive quarter of declining revenue. Not good.

We learn that a phenomenon called "disruptive innovation" is the cause of such problems for incumbents in the analytics industry. Disruptive innovation doesn't necessarily happen because of a new technology. It more often happens when a company creates a new business model that disturbs the existing relationships between suppliers, channels and buyers, and puts alternative connections in their place. It's something that comes along and rewrites the rules, shakes up the status quo, and changes the scope of an entire industry.

Companies that don't adapt to this disruption will see their value networks break down and their success take a knock.

Although the examples here are specific to the analytics industry, it's a pertinent lesson to guard against complacency – whatever sector you work in.

As well as causing problems for large, established organizations, disruptive innovation is a source of opportunity for other, more nimble companies, and it's a vital feature of the new business analytics landscape.

So let's hear about some of the innovations covered by this book, which are disrupting business analytics today.

Chapter three looks at open source analytics, or, as the author puts it, "the disruptive power of free." The key takeaways from this chapter are how the wider business community can benefit from free and open source models, and the lessons of how they disrupt established software markets.

To start with, the absence of licensing fees sets free and open source software apart from commercially licensed software. Without fees, there's nothing to stop people from freely trialing the software, and then adopting and using it. And while open source models may not be as rich in features as their commercially available cousins, their "good-enough" functionality allows most users to achieve what they need to achieve without spending a fortune. This gives them a significant edge over established software applications.

Interestingly, though, we learn that there's more to it than just the lack of a fee. Open source software is also "free" in the sense that its developers share the source code and allow users to modify and redistribute it.

This opens the door to the second way in which open source software disrupts industry incumbents. There's a community around open source software, where users can innovate and donate software enhancements back to the developer. This sharing of code means that there are no barriers to innovation from other parties. With more input from end users, open source programs develop at a faster pace than commercially licensed products. The most innovative techniques therefore tend to appear first in open source software.

Perhaps the most eye-opening thing we learn in this chapter is just how pervasive open source software has become within the analytics world. It's a growing feature of every software category. Hadoop, Spark, Flink, and TensorFlow are just a few examples of how open source software has seriously disrupted the industry.

Further on in the book, chapter seven looks at analytics in the cloud, and the disruptive power of "elastic computing." The term "cloud" refers to internet-based infrastructure, software and platform hosting services, and, according to the author, cloud computing is taking over the IT world. It's disrupting the technology and computer hardware industries, as well as the success of the leading business analytics providers.

The author predicts that worldwide revenue generated on the public cloud will likely grow from nearly $70 billion in 2015 to more than $141 billion in 2019.

Cloud computing is gaining ground for two reasons. First, there's a greater variety of software options in the cloud, meaning that organizations have more to choose from, and they can switch from one to another quite easily. Second, the cloud allows companies to benefit from what's called "elastic computing" – the idea that customers should only pay for what they use. This has obvious benefits.

We learn that companies that provide their technical services in the cloud benefit from economies of scale and skill that IT organizations within individual firms would struggle to achieve on their own. In the cloud, they're better equipped to cope with peak workloads and sudden surges in demand, more able to scale up their operations very quickly, and less averse to trialing pilot projects. These, and a whole host of other gains, are leading more and more firms to move their operations to the cloud.

After hearing the business case for cloud computing, you reach a section on "Analytics in the Public Cloud," which we particularly like. Here, the author demonstrates how it's now possible to build complete business analytics platforms solely by using the services of the three main public cloud computing providers.

Amazon Web Services, Microsoft Azure and the Google Cloud Platform are these three main players, and Dinsmore walks us through their main services. Storage, databases, business intelligence, machine learning, and marketplace services are amongst those covered, allowing you to make direct comparisons between all three providers and assess which one might be best suited to your needs.

The last chapter – Chapter 10 – might just be the most important part of the book. After nine chapters that have covered a range of different innovations, you'll likely be convinced about the power of disruption. But this final chapter is a different beast. It's a brilliant call to action, a clear, energizing manual for turning theory into practice. It surveys the people, processes, platforms, and tools you'll need in place if you want your organization to benefit from disruption in the analytics marketplace. In short, it's a sequence of commands and rules which, if you follow them, will help you to build your own platforms for disruptive analytics.

So if you read only one chapter in the book, make it this one.

Zooming in on one key point, Dinsmore says it's essential to understand your clients' analytic needs if you're to survive in a disrupted economy. This is especially true if your main business is providing data analytics, but it has a broader relevance too. It's useful to be aware of the different types of data analyzes, and who might be most interested in them.

We learn how C-level executives, for example, require analyzes to dive deep, that are ad hoc and usually quite urgent, and which are highly professional. For managers at an operational level, however, it's real-time metrics that matter most.

Further on, the author looks at the importance of having a "lean data strategy." Around the world, people tend to rely on small amounts of the total data available to inform them when they're making decisions. Dinsmore suggests this might be because not all data is very useful. We like this idea! By his own admission, saying that not all data is useful seems heretical, when we're living in the era of Big Data and cheap storage. The drive to acquire ever more data seems so irresistible and relentless, why would we want to consider acquiring less?

But stop and think for a minute – it's not actually such a radical idea. If you collect data that you're not going to use, that's never going to produce any useful insights for you, then it has no value. Even if you collect data that you think you might need one day, it has no actual value until you come to use it. It's just sitting there, taking up memory space.

Dinsmore's answer to this dilemma is the "lean data strategy." It's the idea of only collecting data for which you have a clear business need, that you can use to deliver genuine insight. This might feel like a radical idea but delivering insight is, after all, the purpose of data analytics, and collecting unnecessary data is only going to make that job more difficult.

So what's our last word on "Disruptive Analytics?"

Well, its overarching lesson is crystal clear. To survive in the business analytics marketplace, you have to disrupt, or you'll be disrupted. It's as simple as that, and it's a truth that could apply in many other industry sectors as well. The book lays out a roadmap to help you be disruptive by aligning your people, your processes and your technology. Dinsmore is clearly an expert in his field, and he peppers his book with case studies and examples that help readers make sense of all the theory.

Overall, we highly recommend "Disruptive Analytics" for anyone who needs to bring themselves up to date with the latest trends affecting analytics and business intelligence. That said, our recommendation does come with a caveat.

This book isn't light bedtime reading for anyone. It's dense, dry and technical, and if you don't come to it with a certain level of understanding about the topic, you'll likely find it a challenging read. If you're new to the subject, you may have heard about "big data" and perhaps "data warehousing." But concepts like "embarrassingly parallel," "gradient boosted trees," and "latent Dirichlet allocation" might well have eluded you until now. If you're nodding your head in agreement, then we recommend you start with a more general book on this topic.

"Disruptive Analytics," by Thomas W. Dinsmore, is published by Apress.

That's the end of this episode of Book Insights. Thanks for listening.