With the opportunities AI has opened, businesses are taking advantage of the AI momentum sweeping every industry. They leverage AI to improve operations and enhance customer experience. You may be reaping the benefits as a consumer, as Netflix employs AI to offer personalized recommendations, heightening customer engagement and subscriber retention. 

Similarly, Google’s AI integration into its search engine algorithms, Uber’s use of AI algorithm to refine its pricing strategy, and Amazon’s use of AI to elevate its recommendation system have ushered in a new era of search precision.

As we delve into the intricacies of each application, it becomes evident that AI is enhancing operational efficiency and shaping a more personalized customer experience, which may translate to increased revenues for these industry leaders. 

While they reap AI benefits, what are the fundamental factors organizations must consider to ensure a good return on investment (ROI) from their AI investment?

MEASURING AI’s ROI 

Calculating ROI from AI initiatives poses a formidable challenge. Unlike traditional investments with straightforward cost and benefit structures, the challenge arises from the interplay of various factors within AI. 

Why is measuring ROI difficult? 

“The first reason is complexity. AI implementations can be intricate, involving various components and stages, making it challenging to accurately quantify the overall impact on investment. The second is time lag; the returns from AI investments may not be immediate, as it often takes time for the technology to integrate, learn, and start providing tangible benefits,” says Sid Bhatia, Regional VP & General Manager, Middle East, Turkey and Africa at Dataiku, an AI and machine learning company. 

On top of these, he adds various variables, such as evolving technology, changing market conditions, and unforeseen challenges, that can introduce uncertainty into ROI calculations for AI projects.

A recent  PwC report that aims to address AI’s ROI problem found that while businesses are reaping the benefits of AI, they’re not often seeing a financial return or not even covering their investments. 

The report stated that the compounding challenge facing organizations is the struggle to define ROI for AI in the first place.

According to Jessica Constantinidis, Field Innovation Officer, EMEA at ServiceNow, a cloud computing platform developer, understanding the ROI in AI implementation is closely tied to a comprehensive comprehension of administrative costs associated with various processes.

She points out a prevalent challenge wherein many companies lack a precise understanding of the complete costs involved in a process, encompassing tasks like finding data, data entry, and data validation. 

Constantinidis highlights that these steps are often overlooked in cost calculations. The commonly adopted approach involves considering the salary per hour of an employee tasked with finding data and multiplying it by the average time it takes, roughly three hours per week, to arrive at the cost of the process.

However, she emphasizes that this calculation does not account for the ‘hidden’ costs previously alluded to, including administrative steps beyond the basic search for data. Furthermore, she underscores the oversight of additional efficiencies gained through AI implementation, such as faster time to market, quicker issue resolution, and intangibles like improved employee and customer retention.

“Calculating ROI from AI is not as hard as you might think. The two biggest factors are how long it takes to build and how long it takes to train. You can measure the cost of both of those, usually at a cost per hour for labor or computing power. Once it’s built, it gets fun because AI can save you millions of dollars. In our case, we’ve replicated the work of an entire newsroom with a simple click of a button,” Muhammad Lila, Founder & CEO of Goodable, says.

Ironically, Constantinidis says, “Using AI (machine learning and process optimization), organizations could calculate the difference in the cost of using AI versus non-AI, so the more transparency they get, the more detailed info they can get.” 

SOFT ROI vs. HARD ROI 

What if the actual benefits of AI extend beyond immediate ROI calculations? Beyond the tangible gains, what about the intangible returns that contribute significantly to an organization’s long-term success? 

“This is why it is imperative to understand where the budget for AI projects comes from. AI should be a corporate-wide budget calculation and not just an IT calculation since a lot of the ROI of AI is intangible and shows up in terms of things like better customer experience and employee satisfaction,” says Constantinidis. 

According to Bhatia, three critical areas exemplify the challenges in measuring AI’s impact. Firstly, in the customer experience arena, AI has demonstrated its effectiveness through personalized interactions, rapid issue resolution, and anticipatory services, all contributing to heightened customer satisfaction and loyalty. Bhatia says the difficulty in “assigning a precise value” to these enhancements. 

Secondly, innovation and adaptability emerge as key domains where AI excels by automating processes, enabling businesses to navigate market changes more efficiently and potentially gaining a competitive edge. The inherent challenge is quantifying the innovative advantage brought about by AI. 

Lastly, employee productivity sees a significant boost through the improved efficiency and automation facilitated by AI, allowing employees to redirect their focus toward more strategic and creative tasks. 

However, Bhatia points out that determining the exact ROI in this area proves to be elusive. These critical domains underscore the intricate nature of evaluating AI’s impact, where its true value often resides in the intangible and multifaceted realms of customer satisfaction, innovation, and employee productivity.

“Think of how much time companies spend looking for sales leads, building new strategies, and other tasks that take a long time. AI can automate a lot of that process. That’s intangible because it removes layers and layers of planning, stress, and mental space that founders can use for other things,” Lila says.

Experts agree that a comprehensive review should extend beyond the tangible cash and financial value of hard investments in building AI projects when evaluating your company’s AI expenditures. 

It is imperative to recognize and prioritize the soft investments – employee training, cultural shifts, and the time invested in understanding and integrating AI into existing workflows – that play a pivotal role in ensuring a robust return on AI investments. 

THE INTANGIBILITY OF AI RETURNS 

Achieving optimal returns necessitates a holistic assessment that values both the concrete financial aspects and the softer yet equally vital elements that contribute to AI initiatives’ overall success and sustainability.

“It really boils down to identifying specific use cases that lend themselves to AI. If the use case is a case for AI, not just a Google query, the ROI will bear itself out over time and be significant. But unless someone finds that massive use case, most AI use cases are very costly as we still have much to learn about prompting and using AI properly in an organization,” says Constantinidis.

According to Bhatia, the positive experiences of loyal customers, facilitated by AI-driven solutions, contribute to organic marketing through word-of-mouth recommendations and referrals. 

This has a knock-on positive impact on brand reputation, potentially increasing market share.

 “And while immediate ROI might not capture these intangibles, the long-term value of a loyal customer base and positive brand image can significantly contribute to sustained business success and profitability. Incorporating customer lifetime value considerations into ROI calculations can provide a more comprehensive view,” adds Bhatia. 

As organizations navigate the AI landscape, understanding and acknowledging intangible yet impactful returns becomes essential for comprehensively evaluating the technology’s overall impact and value.