Seven essential questions leaders must ask before engaging with AI
11 Mar 2024
Blog
Artificial intelligence promises extraordinary performance gains, but many initial forays by businesses falter, wasting time, money, and effort. Worse, when AI doesn’t work or adds to issues, customer and employee experience declines, and workforce morale sinks. Engaging properly with AI at the outset, then, is arguably business-critical.
“Most AI projects fail,” writes Iavor Bojinov, a former Data Scientist at LinkedIn, in the November-December 2023 edition of Harvard Business Review. “Some estimates place the failure rate as high as 80% – almost double the rate of corporate IT project failures a decade ago.”
Why? Business leaders fall prey to the hype, seeking silver bullets while missing the target of prudent scrutiny. Before adoption, probing assessment matters on organizational readiness, data capabilities, value alignment, and more.
Here are 7 questions leaders should ask before committing to an AI program.
1. Are you fixing a problem that’s broken rather than pursuing new value?
Scrutinize targeted processes objectively. Reconsider if you are attempting to bandage dysfunction with AI rather than unlock new opportunities because AI won’t magically fix the reasons current processes often underperform. The people, policies, and leadership behaviors yielding poor results will likely integrate dysfunction faster through automation.
Instead, identify hitherto impossible capabilities AI now makes achievable. Fix chronic issues first, then use AI for innovation, not rehabilitation.
2. Does your strategy, culture, and structure allow integrated, data-based decisions?
AI built on fragmented data and siloed decisions will automate dysfunction faster. Assess if your structure encourages unified perspectives and governance through facts, not opinions. If not, address this first.
Active governance matters tremendously. Many leaders take a curiously hands-off approach to AI, ceding critical decisions to IT departments. However, given AI’s vast business transformation potential, leaders must grab the steering wheel.
AI should act as a co-pilot, empowering leaders with data-driven insights while relying on human judgment for the most significant decisions
Leaders must own the strategy, culture, and operational transformations necessary for AI’s success. Don’t expect it to mend impairments magically. Approach it with optimism but guard against overhyping and unrealistic expectations.
3. How exactly does AI target valuable new opportunities?
As mentioned above, don’t fix chronic issues with AI. First, address why performance suffers, then apply AI for unprecedented innovation, not rehabilitation. Ensure a compelling business case targeting new value. The foundation for this approach must be a robust data strategy.
Notably, Gartner’s Top Strategic Predictions for 2024 and Beyond, published in December, warned: “While GenAI brings a great deal of opportunity, ‘malinformation’ is a new threat vector.” It predicted that by 2028, business spending dedicated to battling this false data would surpass $30 billion, cannibalizing “10% of marketing and cybersecurity budgets”.
There will be other unknown unknowns. Those without the necessary data foundation should not try to build on top with AI.
4. Have you assessed data availability, reliability, and integration?
Data feeds AI’s intelligence. Review its robustness across the value chain. Assess accessibility, quality, and connectivity. Identify gaps obstructing enterprise insights before projecting capabilities or returns.
5. Have you validated measures that capture value rather than just activity?
Many leaders default to activity-based metrics missing economic impact – pressure-test assumptions linking AI projects to financial returns and strategic goals. Ensure clarity on assessed outcomes.
6. Have you pressure-tested unintended consequences like service declines?
Evaluate alignment to customer experience carefully. Will service suffer from over-automation or lost empathy? AI should enhance human judgment, not fully replace it.
7. Have you scrutinized business case assumptions before significant investments?
AI demands significant executive sponsorship, multi-disciplinary skills, and enterprise coordination. Validate business cases before launching complex programs. Here, it pays to have the C-suite educated on the possibilities of AI. CEOs would be wise to train up in 2024 to increase the success of AI-powered innovation – and, moreover, work out whether it is needed or not.
The Harvard Business Review article mentioned above recommends creating a center of excellence “to build an easy-to-use AI factory and provide … employees with tool-specific training and education”. Leaders should be enrolled and actively engaged, too.
Asking these 7 questions will help leaders adopt AI prudently, not rashly, and avoid being oversold. Separate hype from reality. Pressure-test dreams against hard truths. This undertaking remains substantial with few guarantees, so prepare thoroughly and proceed deliberately.
AI can deliver extraordinary gains only if built on reliable data, trust in automation, and leadership capable of navigating uncertainty and complex system dynamics. Start your assessment with the above, then create an adoption roadmap matching operational maturity to technology capabilities.
Finally, leadership teams must govern AI actively, not take a technology backseat. Though promising, it remains largely unproven without comprehensive diligence. Those establishing integrated foundations and pragmatic expectations will more likely thrive in the age of analytics.
Author(s)
-
