How to Compute ROI on AI Projects

Artificial Intelligence (AI) is revolutionising industries, offering transformative potential and competitive advantages where it previously never existed. If you are a business considering AI adoption, evaluating or getting an idea about the return on investment (ROI) will become a defining factor in deciding on whether to go ahead with the project or not. This article highlights different methods of computing ROI for AI projects and what factors one should consider along the way.

ROI measures the profitability of an investment relative to its cost. For AI projects, ROI assesses benefits against initial investment, considering both tangible and intangible returns. However, you need to be aware that when it comes to looking at the ROI for an AI driven project, it is somewhat different from a normal software or IT driven project. This is because AI is by nature an experimental approach that matures over time to create real value and not everything can be measured in monetary terms.

When calculating ROI for AI projects, you should take into account:

Initial Investment: Costs for AI infrastructure, software, data acquisition, and resources.
Operational Costs: Maintenance, training, and support expenses.
Revenue Generation: Potential revenue growth from enhanced productivity, customer experiences, or new opportunities.
Cost Savings: Operational efficiencies and error reduction leading to savings.
Intangible Benefits: Brand reputation, innovation, and competitive advantage.

Various methods are available for computing ROI:

Simple ROI Calculation: ROI=(Net ProfitTotal/ Investment)×100
Payback Period: Time for the AI project to recoup its initial investment
Net Present Value (NPV): Discounting future cash flows
Internal Rate of Return (IRR): Annualised rate considering investment and returns

Which one you choose will depend on the type of AI project you embarking on, the resources you have at your disposal to carry out the project as well as the complexity of the project

Calculating ROI for AI projects presents challenges:

Complexity of AI: Intricate algorithms and evolving technologies
Data Quality and Availability: Obtaining high-quality, available data
Long-term Impact: Quantifying long-term effects like market disruption and scalability

Consider a marketing agency contemplating an AI-driven customer segmentation tool to boost campaign effectiveness and personalisation. They want to calculate their ROI and have evaluated the challenges they are likely to face in order to come to an informed decision. Let’s say they have done all the due diligence through various board meetings and workshops on what they are prepared to spend, what benefits they will be looking to get out of it and the challenges they will need to be prepared for and are now ready to move forward based on the following estimated figures.

Using the simple ROI=(Net ProfitTotal/ Investment)×100 from above, we can apply the following numbers to it:

ROI=(180,000/80,000)×100=225%

To start with, the business will need an AI Strategy in place to help guide them not only to tackle issues such as ROI but also to ensure that they follow a framework that covers all aspects of their business that will be relevant to their AI project.

Based on the above calculations the marketing agency will assess the AI complexity of the project in terms of cost, resources and its objectives as either easy, moderate, or hard. In this case let’s say for argument’s sake they agree on moderate. They will base it on the fact that it will require ongoing technical expertise and updates but it is something they have factored in driven by the projected benefits they are likely to get in the future. This could however increase operational costs by £10,000 annually but their projected ROI will result in faster processing of repetitive tasks that should save them time and money in the long run which in turn will drive down cost. One other thing to consider. If they are looking for short term ROI this may not be the right course of action to take. If however, they are looking for the long term ROI, this maybe the right projection. The key thing here is that there are no straight forward answers and the decision to move forward will be a joint decision between management and the right people involved in the project. Below are a few other areas the marketing agency will need to consider to make that decision.

High-quality data is essential for accurate customer segmentation. The agency may need to invest an additional £5,000 annually to ensure data quality and availability.

The agency will need to anticipate market changes and scalability needs, estimating potential additional costs or revenue opportunities that could impact ROI by ±10%.

ROI of 225%: With a high ROI, the agency sees significant financial potential in the AI project.
Complexity: Moderate AI complexity increases operational costs but is manageable with expected revenue growth.
Data Quality: Ensuring data quality may require additional investment but is crucial for AI effectiveness.
Long-term Impact: Market changes and scalability considerations highlight the need for flexibility and adaptation.

Considering these factors, the marketing agency concludes:

1. AI Implementation Pros: High ROI and revenue potential.
2. Challenges: Manageable operational costs, necessary data investment, and adaptability required for long-term success.

Calculating ROI for AI projects involves assessing initial investments, operational costs, revenue generation, cost savings, and intangible benefits. While challenges exist, informed decision-making through comprehensive evaluation is the key to success.

In our hypothetical example, the marketing agency’s calculated ROI of 225% provides a compelling case for AI adoption. By addressing challenges proactively, the agency can harness AI’s transformative potential, positioning itself for enhanced competitiveness, innovation, and success in a dynamic market landscape based on long term ROI.