New Delhi |
Published: July 13, 2020 1:44:44 pm
Implementing Artificial Intelligence (AI) in an organization is a complex undertaking as it involves bringing together multiple stakeholders and different capabilities. Many companies make the mistake of treating AI as a ‘pure play’ technology implementation project and hence end up encountering many challenges and complexities peculiar to AI. There are three big reasons for increased complexity in an AI program implementation – (1) AI is a ‘portfolio’ based technology (example, comprising sub-categories such as Natural Language Processing (NLP), Natural Language Generation (NLG), Machine Learning) as compared to many ‘standalone’ technology solutions (2) These sub-category technologies (example, NLP) in turn have many different products and tool vendors with their own unique strengths and maturity cycles (3) These sub-category technologies (example, NLG) are ‘specialists’ in their functionality and can solve certain specific problems only (example, NLG technology helps create written texts similar to how a human would create it). Hence, organizations need to do three important things – ‘Define Ambitious and Achievable Success Criteria’, ‘Develop the Right Operating Rhythm’, and ‘Create and Celebrate Success Stories’ to realize the true potential of AI.
Strategy: Define Ambitious and Achievable Success Criteria
Most companies have very narrow or ambiguous ‘success criteria’ definition of their AI program. These success criteria are not defined holistically and hence may end up providing sub-optimal benefits to the organization. We suggest that the success criteria of an AI program need to not only be ambitious, achievable, and actionable but also tightly integrated with the overall key strategic objectives and priorities of the organization. For example, a bank which is trying to reduce the number of customer complaints and improve the customer experience as key strategic goals can benefit immensely from integrating AI program goals with the goals of this important program (example, leverage machine learning and analytics to analyze past complaints data and better understand customer complaints patterns and journeys and ‘decision points’). This interlocking of success criteria will help AI program leaders with the right yardsticks to align and measure their progress and contribution. Additionally, it also helps them get the right visibility and sponsorship at the senior leadership levels in the organization that further improves the chances of success of the AI program.
Operations: Develop the Right Operating Rhythm
A successful AI program requires four key ingredients – Right Data, Diverse Skills, Scalable Infrastructure and Seamless Stakeholder Alignment. It is said that ‘Data’ is the food of an AI program and hence having the right data (example, the volume of data, type of data, and quality of data) at the right time is critical to ensure AI programs have the required fuel and energy to complete their intended journey. While good AI skills are in short supply, leveraging constructs such as having a Nimble CoE (Centers of Expertise) increases chances of optimal utilization of these rare and expensive skills across the organization. Finally, getting various important stakeholders (example, Global Process Owners, IT Leaders, Internal Control & Risks, Continuous Improvement, and HR) seamlessly work together is important to reduce friction and increase AI program velocity.
Change Management: Create and Celebrate Success Stories
It is said that ‘success breeds more success’. While AI programs typically focus a lot on efficiency and productivity improvements, many AI programs also generate significant non-direct-quantifiable benefits (for example, improvement in stakeholder experience, improvement in employee engagement and morale). A recent Deloitte survey indicates that 44 per cent organizations felt AI has increased the quality of their products/services while 35 per cent organizations found that AI has enabled them to make better decisions in their organizations. Successful companies find a way to identify these simple, holistic stories and narrate them compellingly and consistently in multiple forums at all levels in their organizations. Humans, by design, are inspired better by stories (than by just numbers) and hence creating a powerful story that combines the quantifiable (example, number of hours saved) with other benefits (example, better decision making ability) can galvanize the entire organization and facilitate rapid and increased adoption of AI at all levels and in all units of the organization.
The revered Chinese saint, Lao Tzu, once famously remarked that ‘A journey of thousand miles begins with a single step’. The AI journey in an organization is no exception. While AI implementations are typically more complex and nuanced, companies can leverage the 3-pronged approach mentioned above to realize the true and full potential of AI. While a successful AI program implementation can bestow significant financial benefits on an organization, it also activates the divine journey of freeing up humans to do what they do best – leverage their sophisticated brains to introspect, explore, learn, love, empathize and solve the most intricate and defining problems of our generations.
The authors are Ravi Mehta, Partner; Sushant Kumaraswamy, Director; Sudhi. H, Associate Director; and Prashant Kumar, Senior Consultant, Deloitte India.
📣 The Indian Express is now on Telegram. Click here to join our channel (@indianexpress) and stay updated with the latest headlines
© IE Online Media Services Pvt Ltd