Feature
As enterprises deploy more artificial intelligence (AI), their expectations for the technology’s return on investment (ROI) are becoming clearer.
At least 30% of generative AI projects may be “abandoned” by the end of next year after the proof-of-concept phase, according to Gartner. Yet, nearly three out of four senior business leaders who do deploy AI are reporting ROI for three key metrics: operational efficiencies (77%); employee productivity (74%); and customer satisfaction (72%), according to the “EY AI Pulse Survey.”
Here, AI leaders share how enterprises should assess the ROI of AI and when they can expect to see returns from AI.
AI and ROI
- Defining ROI Metrics for AI
- Tracking AI ROI in the Enterprise
- Why the ROI of AI Should be Established
- When to Expect ROI on AI
Defining ROI Metrics for AI
Productivity
The biggest metric for AI ROI is productivity, said Carlos Meléndez, VP of operations at Wovenware.
Meléndez said an enterprise’s AI productivity metrics can vary by department. For instance, for software engineers, he said, a metric can be the lines of code developed. In recruitment, some metrics can be new candidates, emails responded to and interviews conducted.
“It’s all about developing a baseline of productivity and measuring against it,” Meléndez said.
Enterprises might have different definitions of productivity, but “streamlined processes should directly influence operational costs,” said Pedro Portela, head of consulting at Indicium.
Portela said decision-making time is an AI measure, as “less time looking for information translates to less cost.” Enterprises, he said, can use generative AI to significantly reduce the “time to insight.”
Operational Efficiency
Enterprises can set operational efficiency as an ROI metric when they measure the cost savings from using AI to automate repetitive tasks, such as data entry and quality control, said Robin Patra, head of data, platform, product and engineering at ARCO Construction.
Customer Satisfaction
On the customer side, when enterprises use generative AI tools to improve interactions, loyalty and retention, customer satisfaction is an ROI measure, Portela said. Enterprises, he said, can also measure customer acquisition costs and customer lifetime value related to AI.
When AI is tied to an organization’s net promoter score (NPS), customer churn and customer resolution time, customer experience (CX) is an ROI measure, Patra said.
Revenue
Patra added that revenue is a critical ROI metric, as enterprises can measure sales from AI tech, such as personalized recommendations and dynamic pricing.
To evaluate the ROI of AI, enterprises should start with pilots to “demonstrate value” before scaling and use several key metrics, such as savings, revenue and risk reduction, said Vijay Kotu, chief analytics officer at ServiceNow.
Value Over ROI
However, Kaj van de Loo, chief product and technology officer and chief innovation officer at UserTesting, said “looking for a hard business case for AI is misguided.”
“Do you have a business case for email? For electricity?” van de Loo said. “AI will become part of how we do business. Period. Same goes for trying to measure ROI. Don't. Focus on driving value.”
Related Article: 5 AI Metrics Every Leader Should Track
Tracking AI ROI in the Enterprise
Enterprises should begin tracking AI ROI by “clearly defining the business problem AI will address” and linking it to measurable goals, Kotu said.
Employee Functions
Enterprises can monitor the efficiency created by using AI in specific departments, such as operations, finance and HR, Patra said. For instance, HR could track the length of hiring cycles.
IT and human resources are “gaining the most ROI from AI, but with the rise of LLMs, there are productivity gains for everyone,” Meléndez said.
For technical guidance, a data product manager can “help define which metrics capture the value users are extracting from the solution,” and an AI engineer can provide the technical expertise to extract usage metrics, Portela said.
Organizationally, an enterprise can track productivity from AI by looking at “value calculations” for key roles, such as agents, process owners, developers, end users and leadership, Kotu said.
Customer Behaviors
Customer-related AI ROI can be determined by gathering customer feedback before and after the use of AI, Meléndez said. For example, he said, in a call center that has implemented chatbots or AI agents, an enterprise can survey customers to determine their satisfaction levels.
Enterprises can analyze customer data and preferences and track ROI after employing AI to personalize marketing campaigns, product recommendations and customer experiences, Portela said.
Organizations can also track AI ROI by examining changes in purchase frequency and average order value due to AI-driven personalization, Patra said.
Internal Processes
Internally, the key to tracking AI ROI is establishing baselines of workforce productivity, and it’s important to “take the pulse of employee satisfaction,” Meléndez said.
“By gathering employee feedback, measuring and benchmarking it, you can determine if AI is making employees’ jobs better and if they are feeling valued by the organization,” Meléndez said.
A “success case” of internal ROI is tracking how AI impacts information retrieval and reduces the time spent looking for information, Portela said.
Enterprises can also track AI ROI throughout their supply chain operations, such as predictive maintenance and reducing equipment downtime, Patra said.
Related Article: Where’s the Generative AI ROI? Start With the Supply Chain
Webinar
Sep
25
CMS Briefing: A Live Look at What’s Next in AI-Driven Platforms
Learn how leading organizations are using AI‑driven tools to publish faster, personalize smarter and stay secure.
Register
Webinar
Oct
15
Agentic AI Playbook Real-World Customer Service Use Cases You Can Deploy Now
Boost self-service by 30% and slash call volume by 63% with agentic AI.
Register
Webinar
On demand
Ready or Not: How Data-First Organizations Are Unlocking Agentforce Potential
Learn how to cut through the noise, activate Agentforce and build a Salesforce AI strategy that actually delivers.
Watch Now
Webinar
On demand
AI in Customer Service: Faster Resolutions, Happier Customers
Don’t let rising demand burn out your team. See how to build a smarter, more resilient support org.
Watch Now
Webinar
On demand
From Hype to High-Impact CX Strategies That Actually Scale
Turn buzzworthy AI and outsourcing trends into measurable CX wins with fresh 2025 data.
Watch Now
Webinar
On demand
Insights to Action Rethinking the Contact Center for Real Business Impact
Join our exclusive webinar to hear CX executives share their innovative strategies for transforming service delivery.
Watch Now
Why the ROI of AI Should be Established
Enterprises need to measure and quantify the results of AI just as they do for any other IT investment, Meléndez said. AI, he said, can be “a costly investment, so it’s important to have this quantifiable data to share with CFOs and those holding the purse strings.”
Yet, if AI ROI is not proven shortly after implementation, “it doesn’t mean the organization won’t benefit from AI — it just may not be the right tool at this moment,” Meléndez said.
AI isn’t only a technical deployment, it’s a “strategic lever,” Patra said.
“Enterprises that measure, track and act on AI ROI will unlock sustained competitive advantages,” Patra said.
“Without clear ROI, AI initiatives risk becoming costly experiments rather than value drivers.”
When an enterprise establishes AI ROI, it ensures that AI investments focus on business-critical areas, budgets are allocated more effectively, high-ROI projects are scaled and gaps are revealed to allow for recalibration, including with AI models and re-directing investments, Patra said.
AI solutions that can’t prove ROI or at least the expectation of ROI will “eventually be labeled as tech fads and will most likely be discarded,” Portela said.
Portela said establishing, tracking and continuously evaluating AI ROI “ensures strategic impact and long-term funding.”
Related Article: How the New AI Math Challenges Customer Experience ROI
When to Expect ROI on AI
While it takes time to deploy AI and get teams familiar with the technology, in three to six months, an enterprise should “expect to see some changes,” Meléndez said.
“It doesn’t mean it will be 100% there, but they should see if productivity is improving,” Meléndez said. “If not, they may need to re-evaluate and adjust.”
The timeline for AI ROI depends on the complexity of the initiative, Patra said.
For instance, in three to six months, Patra said, an enterprise can see ROI from using AI to automate routine processes and deploy chatbots.
In six to 12 months, he said, AI use cases such as dynamic pricing and predictive analytics can “deliver measurable business impact after sufficient data is gathered.”
In over a year, more strategic transformations, such as AI-driven R&D and large-scale supply chain optimization, may begin to yield significant value, Patra said.
Kotu added that while it may be tempting for enterprises to take a wait-and-see approach on AI as they for the AI market to mature and AI’s value to become self-evident, “now is the time” to apply, measure and improve AI.
Related Article: MIT’s 4 Stages of Enterprise AI Maturity
