Measuring the ROI from AI Investments

Executive Summary

By now, most of the organizations you know or your own organization is trying out AI use cases to various degrees. AI is an evolving space, with the promise similar to that offered by internet and e-Commerce in early 2000s. But have the decision makers of these organizations thought of what and how to measure the ROI on these use cases?

AI Applications

Organizations have quickly warmed up to the great potential offered by AI. In the consulting world, i am already seeing our clients try out both internal and customer facing use cases with AI. The adoption curve has been growing exponentially.

In my perspective, probably driven by the trust in digital transformation necessitated during COVID-19 or after seeing the impact of digital first channels, companies have been bold in trying out AI in what were traditionally considered “critical” use cases. Customer service, communications, risk management, etc., were traditionally considered “critical” or “sacrosanct” because of their impact on customer satisfaction and customer’s impression of a firm. But in case of AI, Fortune 500 firms have been quick to try AI even in those critical areas as well.

Experimenting with Data

Fortune 500 firms, with their large revenue base, often experimented on the side while staying “traditional” in their digital adoption in critical value chains. Only tried and tested digital tools were deployed in those critical value chains.

Financial Services and Healthcare firms, that deal with HIPPA and PII data, were very conservative in terms of applying emerging technologies to their proprietary data. But the arrival of ChatGPT has changed all that. The potential to monetize a resource which they already had, proprietary data, seemed too good of an opportunity to pass up.

Next Stage - Measuring the ROI

Decision makers across industries were right in quickly adopting AI to various use cases. In a game where data is the resource, it pays to be the first mover. But as these decision makers go back to their boards for next stage of funding, they are sure to face questions on ROI from these investments. How can the decision makers prepare to answer these questions and continue to get funding, in a year of uncertainty due to (still high) interest rates and moderating but stubborn inflation?

It makes sense to measure the ROI on these use cases from the following lens:

Impact on revenue: How has AI enabled our revenue generating lines of business? If not a direct impact on top line, has AI built a solid foundation for the key lines of businesses to do more in the coming year? Have we learnt more about our key customer segment? Have we increased customer awareness about our products and services? If yes, by how much? Has AI helped us to better allocate our resources to grow our revenue? Can we build new sources of revenue using AI?

Cost Savings: How have our automated processes reduced our operating costs? Have AI use cases enabled our business units to service more customer segments? Have our cycle times gone down? Has AI shortened our employee learning curve? How best can we allocate people to new roles enabled by AI?

Efficiency gains: This goes hand in hand with cost savings lens. But this is THE key potential of AI - increased productivity. How productive has our business unit/organization become post AI implementation? Can we quantify it? More importantly, can we scale it across business units? Are there any roadblocks to scaling it?

Customer Satisfaction: Have our customers noticed a difference in our service post roll out of AI? Have our customer queries being resolved faster? Have our CSAT scores started to go up? While AI is still in the early stages of adoption to make a meaningful impact to CSAT scores, what are our vocal customers saying about our service? What feedback have they provided us?

Closing Comments

Finding answers to the above questions or at least having a plan to get these answers will better prepare the decision makers when they go in front of the board. AI is still in early stages of adoption but it is not too early to start to measuring it’s impact on business. By building a business case with a detailed plan to measure the ROI, the decision makers are more likely to win budget for sustained investments in AI. Are you in a similar boat? Let me know in the comments or by messaging me.