2 days Ago
Jo Debecker is Managing Partner and Global Head of Wipro FullStride Cloud. With AI investments accelerating, businesses face a crucial challenge: understanding the ROI of those investments. In fact, this inability to measure the impact of AI is preventing enterprises from tapping into AI's full potential.
Why is it so difficult for businesses to measure the return on their AI investments? The ROI Challenge Of AI One of the primary obstacles is a lack of well-defined goals for AI initiatives. Without clear objectives, it's challenging to link AI outcomes to specific business values, such as cost savings or revenue growth. For example, a company might implement an AI system to improve customer service without setting specific targets for improvement.
Without targets, measuring the system's impact on customer satisfaction or response times becomes difficult. KPIs are also critical. A business might deploy AI to enhance operational efficiency but fail to establish KPIs like process completion time or error rate reduction, making it hard to assess the AI's effectiveness.
Another challenge is poor data quality or inadequate data tracking, which can obscure the understanding of AI initiatives. Take, for instance, an AI solution to enhance inventory management within a retail operation: The system's value would be measured through outcomes such as lower inventory costs, faster stock turnover or fewer sales lost due to out-of-stock situations. Gathering detailed data on these outcomes can be challenging.
For one, it requires comprehensive IT infrastructure to monitor and gather data across various supply chain segments, which not all organizations have. External influences like seasonal demand shifts and supply chain interruptions can also complicate data analysis. The AI system's performance might also adjust over time as it processes new data, requiring a dynamic approach to data collection to maintain accuracy.
All of this makes it hard to directly attribute financial improvements or efficiency gains to the AI project. Without solid evidence of the benefits, convincing stakeholders to continue or expand their investment in AI technology becomes a challenge. How to Measure And Maximize Return On AI Investments A multi-faceted approach can help businesses overcome these hurdles, enhancing the accuracy of their ROI assessments and strengthening stakeholder confidence in AI investments.
Here is what the approach can look like: 1. Invest in a robust IT infrastructure. More than just purchasing the latest hardware, this means integrating systems that ensure seamless data flow across all segments of the supply chain.
Such integration allows for real-time monitoring and data collection, providing a comprehensive view of an AI system's impact. 2. Establish practical steps for improving data collection capabilities.
This starts with modular upgrades or cloud-based solutions. Cloud-based AI platforms can allow for flexible scaling as the organization’s AI needs to grow. These platforms also often offer tools, models and infrastructure to experiment and deploy AI without heavy upfront investments in hardware.
Along the way, though, you'll need to understand cloud economics, which can provide a framework for AI spending, including how to grow promising AI programs and cost-effectively sunsetting AI programs that underperform. 3. Develop a framework for data analysis that accounts for external variables affecting performance metrics.
This involves setting up control groups or using statistical models to isolate the impact of the AI system from other factors, such as market trends or operational changes. 4. Adopt a dynamic approach to data collection.
As AI systems learn and adapt over time, the parameters for data collection might need adjustments to reflect changes in performance accurately. Regular audits of data-collection processes and performance metrics can help ensure that the data remains relevant and accurately reflects the system's impact. This also involves training teams to recognize and adapt to these changes.
5. Be upfront with stakeholders about the challenges and limitations of measuring AI's impact. By openly discussing the steps taken to address these challenges and the methodologies used for calculating ROI, businesses can build trust and set realistic expectations.
This transparency, combined with clear documentation of the AI system's performance and its contribution to financial or efficiency gains, can significantly bolster stakeholder confidence. By tackling these challenges head-on, businesses will not only improve their ability to measure the ROI of their AI initiatives accurately but also enhance the overall effectiveness and reliability of their AI systems. This paves the way for more informed decision-making and stronger justification for future investments in AI technology.
Of course, enterprises do need to consider the possibility that otherwise value-additive AI initiatives can quickly become liabilities if teams rush ahead without proper guardrails. Hallucinations, data breaches and biased outcomes can lead to significant reputational harm. Responsible AI practices—such as privacy and security by design, fairness in processing and outputs and safeguarding of copyrighted information—should not be overlooked as key elements of achieving ROI over the long term.
Managing Board-Level Expectations Managing board-level expectations around AI implementation is another layer to the problem of demonstrating ROI, and doing so requires a pragmatic approach. Providing real-time data and predictive insights can help board members understand AI's impact and make informed decisions. Emphasizing the competitive necessity of AI can motivate action while highlighting AI governance and trust can ensure scalable deployment.
For example, a financial services firm implementing AI for fraud detection should provide the board with real-time data on fraud prevention rates and predictive insights on potential future threats. This approach helps validate the AI initiative's value and aligns it with the organization's broader goals. Establishing A New Dimension Of AI ROI Measuring and maximizing AI ROI demands a clear, structured approach that addresses the unique challenges of AI implementation.
It also requires organizations to think differently. Because AI spans so many use cases and functional roles, it brings new complexities to managing business expectations. By setting well-defined goals, establishing measurable KPIs, running pilot programs and managing board-level expectations with real-time data, businesses can navigate these challenges and unlock the full potential of AI technologies.
Implementing continuous monitoring, measuring long-term impact and establishing feedback loops are essential steps in ensuring AI initiatives deliver expected results. Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?.
