July 5, 2024
By R "Ray" Wang, founder, chairman, and principal analyst, Constellation Research

Meet the New ROI for AI Projects: Return on Transformation Investment (RTI)

Many organizations perform classical cost-benefit analyses to determine the impact of business technology projects. Although these approaches account for quantifiable metrics, they often fail to capture key attributes such as probability of success and level of difficulty in project type.

Today’s artificial intelligence (AI) and business transformation projects require a much more holistic approach when evaluating the value created for an entire organization. Consequently, boards and their executives seek richer attributes to augment their traditional decision-making techniques, which is where a process dubbed return on transformation investment (RTI) comes in. Four elements drive the RTI methodology: cost, benefit, probability, and project type. Constellation recommends forecasting each element over a five-year period.

RTI is necessary because the usual understanding of economic impact is not enough to determine successful transformation. In conversations with more than 200 leading business transformation executives, Constellation found that most believe existing techniques fail to provide a holistic and pragmatic perspective for a few reasons:

The long view is often ignored in short-term project execution. Even though business transformation projects have longer time frames, most business leaders and technology executives work from yearly budgets but live quarter to quarter. With an average tenure of 2.8 years for transformation leaders, the incentives are not in alignment with long-term benefits. In fact, most ROI calculations fail to consider the longer-term impact analysis.

The likelihood of success is not considered in investment analysis. Executives often make their case for transformation using the most complex analysis and management-consulting types of projections. Unfortunately, project sponsors rarely rank the likelihood of success for these transformation efforts. A proper portfolio analysis requires an understanding of a project’s propensity for success.

Impact analysis fails to consider qualitative attributes. Organizations can easily ascertain hard metrics such as cost and benefit. However, the determination of probability and impact requires industry expertise and access to the best practices from early adopters who can provide validation.

Project types do not reflect the level of difficulty that must be overcome. Cost takeout projects differ greatly from business model transformation. Revenue and growth initiatives are usually more complicated than operational efficiency. Compliance projects will always be funded despite economic status. Applying a difficulty factor can help organizations with prioritization of budget, resources, and time.

The RTI methodology provides a proven approach to evaluating investment prioritization for a portfolio of transformation initiatives. Constellation recommends the following:

Apply a consistent formal methodology to assess transformation efforts. C-level change agents agree that a formal methodology such as RTI will allow for consistent evaluation of projects and assist with prioritization. Formalizing the methodology ensures a consistent approach to quantifying and qualifying the value of a transformation initiative.

Be diligent in developing metrics for each element. What is measured will determine what the results are. Focus on first principles and the metrics that matter.

Consider how technology will transform business models. Focus on the business model and then see what technologies will transform the business. Every project should be driven by the business and apply exponential technologies to achieve the desired outcome.

Determine the sustainable impact of a transformation investment. Some projects have a half-life; others provide durability. Prioritize those projects that have cultural buy-in and create enduring impact.

Educate all stakeholders on the RTI methodology. The methodology for RTI should be taught at the level 1 and level 2 tiers of executive management. Use the RTI methodology across business units, finance, IT, and operations.

AI projects often take an all-or-nothing approach. However, from their accumulated experience, Constellation’s Executive Network (CEN) members suggest a gradual and measured approach. Constellation sees five phases to adoption from both a business and cultural point of view:

Augmentation. Organizations begin by finding tasks that benefit from augmenting existing workflows and learning where exceptions are needed.
Acceleration. As more data is taken into consideration and training improves, false positives and false negatives are worked out to accelerate progress.
Automation. Once a level of confidence has been achieved as more data points are assimilated, organizations can act with confidence toward automation.
Agent. Automated agents play a role in improving personalization at scale.
Adviser. Advisers can provide capabilities in prevention and prediction built on the overall accumulated knowledge and insight.

R “Ray” Wang is the author of the new book Everybody Wants to Rule the World: Surviving and Thriving in a World of Digital Giants (HarperCollins Leadership) and founder of Constellation Research.