Adoption of Artificial Intelligence (AI) in business is still in nascent stages. Since it is a technology that is being used for the first time by businesses, the thumb rules are still under formation. Moreover, AI technology can be used in multiple ways – for increasing efficiency and revenue, and for cutting costs. Thus, companies have achieved widely differential results from AI use in business.
According to the McKinsey report, The State of AI in 2020, less than 25% of the survey respondents (who belong to different companies across industries that have adopted AI) admitted experiencing “significant bottom-line impact”. According to Michael Chui, partner at McKinsey Global Institute, San Francisco, this is not surprising, since achieving large-scale impact with AI is still elusive for most organizations because of both – technical challenges and organizational changes that it demands.
However, since the study found that some companies from different sectors succeeded in drawing 20% revenue from using AI, it is possible for every sector to generate good value from AI, provided it is “applied effectively in a repeatable way.”
After studying and analyzing the survey responses, the McKinsey report found that the companies that were able to garner maximum value from AI were following certain practices. Given below are the five sets of practices that enabled these high performing companies to make a real difference to their bottom-line through optimal use of AI:
Strategy
On an average more than 50% respondents stated the following about their company strategy:
- The company has a road-map that prioritizes AI initiatives that are linked to business value creation across organization.
- Their AI vision and strategy is clear and well-defined.
- Senior management is 100% aligned and committed to the company’s AI strategy.
- They use active programs to develop and manage expansive AI ecosystem partnerships.
- Their AI strategy is well-aligned with the company’s broader corporate strategy.
Talent and leadership
40% respondents from high-performing companies said that in their company tech professionals are aided to develop AI skills via tailored curriculum according to role and progress, along with distinct career trajectories. Additionally:
- The company has a credible leader who is empowered drive AI initiatives in collaboration with colleagues across business functions and units.
- AI initiatives are centrally coordinated and balanced with close connectivity to the business’ end users.
- The nature of AI talent needed to support AI initiatives is well understood and accordingly AU talent is recruited and onboarded.
Ways of working
AI adoption requires significant changes in people’s mindsets and ways of working. Companies that are gaining most from AI are:
- Comfortable to take risks when making investment decisions related to AI.
- They use advanced processes for AI deployment and possess a strong framework for AI governance covering model-development and AI related risk management.
- They use design thinking and involve the end user while developing AI tools.
- Their AI development teams follow a standard procedure for building and delivering AI tools.
Data
- Companies use synthetic data for training AI models, if natural data sets are unavailable.
- Internal structured data is rapidly integrated for using in AI initiatives.
- Well-defined governance processes are in place for data related decisions.
- Have implemented scalable internal processes to label AI training data.
- Data quality is adequately checked, and a data dictionary is in place, accessible to all.
- There is a clear data strategy to support and enable AI.
Adoption
Effective adoption of AI across an organization is extremely essential for it to be optimally successful. It is only possible when the entire organization adheres consistently to the execution processes that have been identified as essential for capturing maximum value from AI. Further, these companies:
- Systematically track key performance indicators for measuring the real impact of their AI initiatives.
- Design the AI capabilities for scalability across business units or throughout the company.
- Possess a clear process for implementing AI solutions from pilot projects to full scale production.
- Enact adequate change management to secure AI adoption.
