The Right People, Collaboration, and Use Cases Propel Top Management to Deliver AI-Driven ROI 

DataRobot and other AI leaders recently collaborated with ESI ThoughtLab, the thought leadership arm of Econsult Solutions, to generate a benchmark study of senior executives in 1,200 companies focused on the impact of AI on ROI. The report’s goals included identifying the way top management implements AI successfully. 

The report revealed that AI isn’t just about technology. Companies that hire the best talent, cultivate top in-house AI skills, and develop an overall culture of collaboration are the leading ROI overperformers.  

The numbers support the importance of internal synergy:

• 83% of ROI overperformers develop and acquire the right people, whereas only 9% of underperformers are doing the same.
• ROI overperformers are superior at training and enabling non-data scientists to deploy AI (88% vs. 2% of underachievers). 
• Firms with high ROI are those that coordinate activity between AI experts and business teams (85% vs. 9% of underperformers).
• ROI overperformers dedicate more staff to AI (on average 894 per firm vs. 193 for underperformers). 
• While 77% of underperformers are centralized, 61% of overperformers are decentralized.

ROI Outperformers Leverage Diverse Data Sources and Emerging Technologies

The report shows that firms generating positive, AI-driven ROI are those that ensure their data is ready to be managed through AI platforms and then augment their fundamental capabilities. For example, Zillow, an online real estate marketplace, has leveraged multiple data sources to assemble a huge volume of data on over 110 million U.S. homes, helping them to get a better picture of the digital desires and demands of real estate buyers, sellers, and agents. 

Zillow not only uses machine learning to run millions of statistical models on a daily basis, it uses high-dimensional data, 3D models,  and computer vision to deliver virtual tours. This is particularly pragmatic in the new era of social distancing. Zillow stays organized with AI-driven planning and scheduling features and has a better picture of its users via investments in natural language processing (NLP) and speech recognition software. Zillow exemplifies how firms implementing diverse data types and AI tools at scale push growth and establish a winning reputation. 

Advanced Sensor Data and No-Code AI Tools Fuel Customer Savings and Create AI Ambassadors

While Zillow is using data to monitor housing needs and costs, GE Aviation monitors weather, flight patterns, and airline schedules, and evaluates sensor data from over 65,000 commercial and military aircraft to anticipate maintenance issues before they actually occur.

Jonathan Tudor, Director of Data and Analytics, relied on “no-code” (i.e., self-serve) AI tools to drive GE Aviation’s AI prowess across the entire company. Tudor chose his tools carefully by cooperating with business and functional units  and inviting employees from across GE Aviation to help him determine the best fit. By working with people, not only did they find the best AI tools but the testers also mimicked the behavior of early adopters and actually became in-house brand ambassadors for their respective tools. 

Industrial Giant thyssenkrupp Identifies “Supercritical Importance” of AI to Improve Supply Chain and Operations

The report also explains how a centuries-old conglomerate, thyssenkrupp, has identified the “supercritical importance” of AI to be competitive and take care of customers in the industrial sector. Abhinav Singhal, Chief Strategy Officer for Asia Pacific, believes that AI can improve supply chain and operations, sales and marketing, and support functions, such as HR and finance, by adding value to three crucial components of the manufacturing and metals sector:

• Augmenting machines and components with a blend of AI and IoT
• Implementing remote monitoring and autonomous equipment to maximize service revenue
• Establishing new business models by assisting customers digitalize plants and implement AI tools to boost productivity and performance 

Singhal reports that lowering inefficiencies surrounding logistics, warehousing, and inventory management has unlocked the highest ROI. Finally, he states that humans will never be replaced by AI, but humans that cannot work with AI will be replaced by humans who can. 

Conclusion

Managing and delivering artificial intelligence can be difficult. It requires a foundation of data management and machine learning, while being prepared to actively integrate new data formats, such as computer vision, NLP, and deep learning. Many of these technological  initiatives, however,  will never be put into production without crucial collaboration with people, talent development, and cultivation of top, in-house AI skills. Whether you are an AI leader or in the early stages of your AI journey, download and read the full report to learn more about the best practices that can help you better manage your team, data, and AI solutions.