International Institute for Analytics

Organizations today face an increasingly challenging business environment. Across industries, new companies and nimble competitors are taking advantage of analytics and leveraging the full potential of the internet, disrupting traditional business models and markets.

Cloud computing and open source have caused fundamental changes in analytics infrastructure, enabling the introduction of new technologies such as artificial intelligence and machine learning. The most nimble, innovative companies have quickly taken advantage of these new technologies — and the analytics they enable — to gain a competitive advantage.

Traditional companies are struggling to deal with this complexity and effectively compete on analytics. Though many top executives realize that high-quality data, analytics and AI are critical to the future success of their companies, up to 70 percent of analytics initiatives and projects fail to meet their objectives.

Since the dawn of Wall Street, investors have sought an edge: an advantage they can leverage to beat the market. Gaining some type of edge is the only way to win in the brutally competitive, zero-sum game of capital markets.

Today, that edge can come from insights gleaned from information found in the exabytes of data created in the real world — on websites, from smartphones and sensors, through supply chains, and from countless other types of real-world data. Collectively, these sources of unstructured and large nontraditional data are called “alternative data.”

The life expectancy of large companies is declining more quickly today than ever. The leading reason for this diminished lifespan is in their persistence to combat challenges that do not alter the company’s success or failure — they are tilting at windmills in a way not too dissimilar to Don Quixote fighting imaginary giants. An incredible amount of resource can be wasted fighting the wrong problem; instead, I propose that large companies methodically chase innovation and move to new business models to survive and succeed.

Each year, the International Institute for Analytics ends the year with a look at the latest analytics trends and the most pressing analytics challenges currently facing organizations. Our predictions are based upon our day-to-day work supporting and advising analytics leaders and organizations. We take advantage of the breadth of expertise and cross-industry perspectives we encounter every day from our clients, partners, and members of the IIA expert network. This is our 10th annual look toward the upcoming year, and our annual Predictions and Priority research brief and the associated webinar have become among IIA’s most popular content of the year. This year, we’ve stuck with our approach of augmenting each of our predictions with a specific priority for leaders to focus on as they attempt to address that prediction. As a result, each priority provides specific guidance as to how to best prepare for, and adapt to, its corresponding prediction.

Open-source software (OSS) is inevitable. For most corporations it’s not a question of “will we use OSS?” The right question is “how much OSS do we already have in production?” More than 90% of IT organizations have OSS in their mission-critical systems. Many commercial software products — from vendors such as IBM and Microsoft — also contain open source software.
Herein we explore these major OSS subjects:

1. OSS history and concepts
2. Quick tour of OSS analytics
3. What does OSS really cost?
4. Positives and negatives of OSS analytics
5. Guidance

To set the stage, the analytics organizational structure I’m presenting below pertains to an analytics organization between 60 and 120 people; this is the size that seems to be a sweet spot for an effective and efficient team (large enough to have specialized skill sets, but small enough to effectively demonstrate the benefits of the team). Moreover, I’m presenting such an organizational design in consideration of an analytics effort at an established, traditional corporation, not a digital native. Digital natives will break down differently with more need for data science and data management. With 60 to 120 people, I prefer a centralized organization with P&L Analytics/Ad Hoc Analysis dotted-lined to their business partners.

Fraud is widespread and continues to grow, especially online. It’s a major problem in a variety of industries and government agencies far beyond the familiar areas of financial and retail fraud, where credit card information is compromised, and fraudsters use it for online purchases. The problem worsens as criminals get more organized and technologically sophisticated and operate at greater scale. Large data breaches expose data about millions of people. Fraudsters automate their activities to exploit thousands of accounts at a time. Online fraud is expected to double between 2018 and 2023. Card fraud continues to be a huge problem with global losses estimated at around $32 billion.

The Internet of Things (IoT) has exploded in recent years due to the dropping costs of sensors, network bandwidth, and data storage. What wasn’t economical a decade ago is now compellingly cheap. As a result, sensors are turning up in more and more places and are generating more and more data. The problem is that as always, generating a bunch of data doesn’t by itself provide any value. What provides value are the analytics using that data. In the case of IoT we will call these analytics the Analytics of Things (AoT). This research brief will dive into a number of important considerations when analyzing sensor data. There will be discussions of success stories, governance, security, technology, and the analytical methods that can be applied to IoT data.

Containers are repositories that hold everything required to run a microservice, software process, application, or data analytics program. Each container includes everything required to run the program: executables, binary code, libraries, and configuration files. This could include Python code or Java code, plus dependencies such as Python modules, JAR files (for Java), interpreters, security software, and the secure sockets layer (SSL).

A container should not access anything outside itself; it is self-sufficient and functions outside of the network. Using a software version or piece of code that doesn’t reside within the container causes leakage. The problem with leakage is that anything located on the host operating system—that is, outside the container—is subject to change.

Any effective journey in analytics involves multiple touch points. Multiple stages of understanding. Multiple vantage points through which data is considered, and extracted intelligence scrutinized. Each step we take is designed to help fill in the blanks. Part of this effort involves outlining the structure of the problems we face, while the rest involves identifying strong solutions to those problems.

It’s a bit like putting together a complex puzzle … of an intricate maze … with multiple possible exit points. On top of that, not all the pieces are immediately available, or never will be, or will be blurry at best. And in contrast to regular puzzles, there aren’t really any obvious borders.