Last week, Vertica sponsored the Chief Data & Analytics Officer event in Chicago. As a first-time sponsor, we were cautiously optimistic that we could meet with data and analytics leaders that were, according to event organizer IQPC, “…on the forefront of capitalizing on data and analytics in the enterprise as the volume, availability, and complexity of data continues to increase and evolve.”
The show met up to its billing. Leaders from data-driven organizations (some Vertica customers) – ABB, ADP, BMW, Cox Automotive, Crate and Barrel, GE Digital, and others – shared their challenges and opportunities in managing the insatiable demand for insight from their data consumers, while embracing constant change and mega trends like AI and IoT.
Day One Presentations – BMW on Data-Rich AI Cycles and Kraft Heinz on the Importance of SQL
On the opening day, Jeff Hamilton, Head of Consumer Insights at BMW, took attendees on his “analytics journey,” starting with Nokia, then Microsoft, and ultimately with the German automobile manufacturer through a series of acquisitions. In addition to a strong recommendation to establish true KPIs for the business, Jeff also shared his five steps to AI transformation, based on Erik Trautman’s Virtuous Cycle of AI Products. The “cycle” or workflow is so simple yet powerful – as more products get used, they generate more data, which are fed into Machine Learning models that ultimately improve those products and generate more usage.
Next, Brian Pivar, Senior Director of Data and Analytics at Kraft Heinz, shared his plans for creating and fostering a data-driven culture – starting with only one data engineer in a company of 30,000 employees! His first recommendation was to retrain and build out a team with solid SQL skills – often referred to as the lingua franca of business. It’s an ambitious five-year strategy, but Brian emphasized the importance of balancing quick wins vs. long-term success to prove the importance of data-driven initiatives.
Emerging IoT Use Cases and Embracing Change for Future Analytical Success
We also had an active discussion in our roundtable on Deriving Value from IoT Data with Analytics at Scale. Topics covered everything from edge computing to graph databases to vehicle telematics to even an emerging IoT use case in casinos where staff receive alerts when slot machines are nearly full of cash (talk about a nice problem to have). Net-net – the use cases driven by IoT analytics are endless and projects are getting funded.
In our ten meetings with data-driven companies in the Chicago area, we also uncovered some common themes, all focused on the need to embrace change:
- Proliferation of data siloes – With two large companies merging into one, an industrial manufacturer had concerns with growing data siloes and the need to analyze that data in place without consolidating workloads for one primary data warehouse.
- Netezza migrations – With IBM announcing end of support for Netezza, companies are in the midst of moving off of analytical appliances to more open analytical platforms and, hopefully, don’t get locked in again.
- Machine Learning for everything – Every attendee wanted to discuss Machine Learning best practices and the latest innovative approaches – in a push to make data scientists more productive and kick their predictive analytics initiatives into high gear. One Vertica customer even praised Vertica’s speed because his data scientists would never wait for more than 60 seconds for a query to run.
- Benchmarks, benchmarks, benchmarks – Failed Hadoop vendors. End-of-life analytical appliances. Lackluster cloud analytical services. Nearly every conversation left us with a request for performance benchmarks and life preservers to move on from the wrong choice of analytical platform.
We’re coming up on the fall trade show season, so meet up with us on the road and keep an eye out here for more trip report findings. And don’t forget to mark your calendars for the Vertica Big Data Conference 2020.
Related posts and resources:
Webcast: Geospatial Analytics at Scale with Vertica
Post: Doing In-Database Machine learning 1 – Why Would You Do That?
Post: Vertica Scores Big in Customer Satisfaction