The adoption of artificial intelligence (AI) in the enterprise continues: More than half (58 percent) of respondents to McKinsey & Company’s recent global AI survey say their organizations have embedded at least one AI capability into a process or product in at least one function or business unit, up from 47 percent in 2018. Those increases were reported across all industries. What’s more, nearly a third (30 percent) are using AI in products or processes across multiple business units and functions, McKinsey’s data says.
But, as the McKinsey research and others point out, some organizations are much further along in scaling their AI initiatives.
[ Do you understand the main types of AI? Read also: 5 artificial intelligence (AI) types, defined. ]
8 things successful AI teams do
What are teams succeeding with AI doing that others can emulate to propel their efforts? Here are 8 habits to consider:
1. Have clear strategies
The organizations that McKinsey identified as AI high performers were deliberate about their plans to scale AI and were more likely to have addressed key issues like business alignment and data. Nearly three quarters (72 percent) of respondents from AI high performers said their company’s AI strategy aligns with their corporate strategy, compared with 29 percent of respondents from other companies. Similarly, 65 percent from the high performers report having a clear data strategy that supports and enables AI, compared with only 20 percent from other companies.
[ Get our quick-scan primer on 10 key artificial intelligence terms for IT and business leaders: Cheat sheet: AI glossary. ]
2. Take a multi-disciplinary approach
Being successful with AI programs requires that organizations create working teams with representation from multiple disciplines, says Seth Earley, CEO of Earley Information Science and author of The AI-Powered Enterprise.
Vodafone, for example, tried to build their AI capability by looking for “cognitive engineers.” “The problem is that cognitive engineer is a new job role and there were none on the market,” says Earley. “Instead, they built their own by assembling a team consisting of data scientists and programmers (obviously), but also linguists, information architects, user experience experts, and subject matter experts from the business.”
The particular mix of skills required will vary based on the flavor of AI. “Predictive analytics would not likely require a linguist, for example,” Earley notes.
3. Cast a wide net
“Companies looking to implement AI-enabled solutions need to ensure they aren’t being limited by their own creativity,” says Dan Simion, vice president of AI and analytics at Capgemini. He advises AI teams to think of as many business use cases for a solution as possible. “While there may be examples of AI-enabled use cases that organizations have implemented previously, there are likely additional cases that have never been thought of. If aligned properly with unique business needs, they could immediately solve an organization’s burning issues,” Simion says.
Casting a wide net of use cases can determine how far the new AI-enabled solution might go – and help the organization identify which use cases are going to offer the quickest payback. “If sequenced correctly, the initial use cases can bring immediate ROI, helping to self-fund future use cases within the program as it progresses,” says Simion.
4. Get specific
“Successful AI projects model what users actually need and determine this through actual working sessions with users, observations, and process mapping,” Earley explains. “These need to be specific and testable.”
AI systems built based on generic use cases like “personalizing the customer experience” will not be testable unless they specify the details of the user, the scenario, and exactly what personalized content and a personalized experience looks like, says Early.
Let’s look at four more best practices:
