Let's Talk

CDO Summit Key Takeaways

Architecture & Engineering
Data Management
Predictive Modeling
02 Apr 2019
7 min read

Data Team

Our Wavicle team had a great time sponsoring the CDO Executive Summit in Chicago on December 11th, where we were honored to introduce the keynote speaker on opening day and to meet big-data leaders from all industries. The one-day event, “built by CDOs for CDOs,” focused on best practices and how to optimize organizations’ data to drive businesses forward.

With all things data, what’s old is new again. Many of the hot topics last week were hot 10 years ago: explosive data volumes, emerging data technologies, skills gaps, and insatiable demand for business analytics. The differences today include new sources of data (cloud, on-premise, social, IoT, etc.); a new landscape of emerging data technologies; and a growing interest in artificial intelligence and machine learning.

And with that, I’ve summarized 4 key takeaways from the presentations we heard and the conversations we had at the event last week. They are:

Predictive analytics

If you predicted (wink) that predictive analytics would be a hot topic, you’d be right. This one was huge and the event panelists covered it from multiple angles. The bottom line is that organizations that incorporate predictive analytics into their business goals and objectives end up with stronger strategies and higher returns than organizations that use limited data sets.

Many aspects of a company’s operations are ripe for predictive analytics. The key is to figure out what challenges can be solved by predictive analytics, prioritize your efforts in some way (for example, ease of implementation or potential value), and then get the data ready.

Typical use cases include: marketing campaign optimization, customer service, fraud detection, and forecasting. For example, using industry standards, predictive analytics takes forecasting analysis to the next level. Traditional forecasting looks solely at historical data to determine future sales. Building a predictive model uses historical data along with a multitude of other variables, including real-time data feeds, to determine future sales more accurately.

Companies incorporating predictive analytic capabilities into their organizations should be prepared to deal with issues including data availability and readiness (including real-time data ingestion), data culture, data illiteracy, skills gaps, machine learning, AI, and alignment across business and IT strategies.

What we heard about predictive analytics last week could fill a Hadoop data lake. OK, I’m exaggerating.

Data monetization

As companies improve their ability to capture, store, and analyze massive data volumes in any format, from any source, they’re looking for ways to increase the value of their data. Whether it’s by identifying new products and services, exploring different business models, or packaging and selling the data itself, opportunities to increase revenue and growth abound.

While most organizations recognize they have a wealth of data, not all of them can realize their potential because of technological and cultural challenges often stand in the way. IT organizations tend to focus on the technology challenges, but to show the value of data, IT leaders need to partner with business leaders to identify where it makes sense to explore monetization opportunities and align the appropriate business processes and technology capabilities to make it happen.

Data integration

As far as we’ve come in the wide world of data, many organizations still can’t get enough of it; can’t get it fast enough; and can’t always get it in a usable format. One session cited research that shows more than 80% of analyst time is spent on data collection and only 20% on generating results. Did I already say that what’s old is new again?

Call it the data pipeline, data integration, ETL, ELT, or whatever you want. As data volumes, types, and formats continue to diversify and grow, one of the biggest challenges IT organizations face is generating a reliable data pipeline.

The goal of an ideal data pipeline is to fade into the background; to allow arbitrary data capture, streaming access, and infinite storage, but otherwise to “just work” efficiently. And they must monitor the flow and alert us about issues the instant they occur so we can fix them promptly. If we can provide analysts with organized, usable data from the start, they can spend more time doing the analysis that helps move our businesses forward.

The industry has spawned many new data warehousing technologies to store and process big data – historical and real-time, in the cloud and on-premises, at rest, and in motion – and data integration capabilities are catching up. Many new tools are emerging to eliminate the ETL “bottleneck” that keeps us from delivering this data in the right form at the right pace for business demands. Can they turn around the 80/20 gap? I am optimistic.

The skills gap

Though none of the sessions was specifically focused on it, many of them addressed the data and analytics talent gap as an undertone of the main topic, whether it was analytics, machine learning, artificial intelligence, or big data in general. It’s not surprising, considering the nascency of many of the technologies being used today, as well as the number of companies in the world that are jockeying for big data dominance. Or at least big data mastery.

A 2017 State of the CIO Survey by IDG/CIO reported that 60% of respondents said they were “grappling with skills shortages” (in 2016, only 49% reported the same challenge). More than 40% cited challenges finding talent in the areas of data science, business intelligence, and analytics, specifically.

Meanwhile, two-thirds of IT leaders responding to a Harvey Nash/KPMG CIO Survey said skills shortages are hindering their ability to keep up with the pace of change.

As educational institutions, corporations, and good old hands-on experience catch up with the demand for big data skill sets, organizations will look for creative ways to close the gap. For example, crowdsourcing, partnering with technical institutes and universities, sourcing from staffing firms, or teaming up with systems integration companies like Wavicle.

In summary

All in all, a great conference here in our home town! As we looked across all these informative sessions and entertained individual conversations at our booth, one other theme emerged. As the only systems integration company to sponsor the event last week, we heard a lot about the challenges organizations are experiencing putting all these pieces together. Whether they’re working on their analytics vision and strategy or migrating to new data warehousing platforms, there are a lot of parts to understand and coordinate.

Do you adjust and upgrade existing tools to accommodate evolving needs? Or is a complete overhaul of your systems and technologies called for? What can you do today to generate quick value, while laying a foundation for future growth? What are the business needs and how do you align technology solutions to meet them?

These are common questions and we are here to help answer them. From strategy through execution, our data analytics experts are on the job with large enterprises in a variety of industries. We make it our business to have deep knowledge and experience in the latest data technologies, including how and when to integrate them with traditional/legacy tools. It was a pleasure to speak with many chief data officers at the CDO Executive Summit, and to hear how many inspiring organizations are earning success with their own big data endeavors.