Despite soaring demand, only 10% of AI initiatives deliver significant ROI.
Artificial intelligence (AI) has become a critical driver of business transformation, enabling organizations to optimize processes, make data-driven decisions, and unlock new growth opportunities. Integrating artificial intelligence (AI) into all business areas is crucial for a company to gain or maintain a competitive edge. According to a 2023 McKinsey & Company research, organizations have reported increased revenue and decreased costs in the business functions where they've implemented AI. Additionally, two-thirds of company representatives expect to integrate more AI into their operations in the next few years.
Undoubtedly, AI is a powerful driver of business growth, offering speed, consistency, scalability, and, most importantly, return on investment. However, as AI investments continue to rise, measuring the return on investment (ROI) is crucial to ensure these initiatives deliver real value and justify their resources. Companies that fully embrace AI report a remarkable threefold return on their investments, setting them apart from those still in the pilot stage. Unsurprisingly, 84% of C-suite executives see AI as vital in achieving their growth aspirations. However, there's one area where AI falls short: ROI. Various studies show that the failure rate for digital transformation projects is between 70% and 95%. According to a Wallaroo.AI study conducted with NewtonX, the failure rate for AI projects is also within this range: 90% don’t deliver any substantial ROI, with 50% not even making it past the prototype stage. With the average ROI at 5.9%, research from IBM suggests that certain companies earn a more notable 13% on their AI projects. So, what sets these 13% apart? A study conducted by Wallaroo.AI surveyed 100 US-based AI, data science, and machine learning decision-makers and highlighted three key things successful projects do differently.
Investment is critical—whether in skilled resources, infrastructure, data, or others. Google, Meta, Amazon, and Microsoft—four Big 5 tech companies—have collectively invested $72.8 billion in AI, while Apple reportedly invests $1 billion yearly. The research found that AI leaders intend to double down on this, with 61% planning to at least double their investments within the next three years and scale their models more than fivefold.
From a resource’s perspective, the size of your AI teams matters. 71% have 100 or more staff working on machine learning. Given that the supply of AI specialists is nowhere near enough to meet demand, attracting and retaining such staff is undoubtedly the most challenging part of securing an AI project's success.
From an infrastructure perspective, many businesses believe working in a public cloud environment is less risky than using a purpose-built, in-house machine learning platform. Regardless of their chosen route, 51% of respondents said the number of tools needed to go into production and the effort required to integrate them was also a significant challenge.
Checking if your infrastructure is AI-ready is essential to determining your AI readiness ranking. For more information, refer to the AI Health Check from our previous blog article. This assessment will help you identify shortcomings and create a successful AI adoption roadmap and implementation strategy.
Back to ROI. Measuring ROI is crucial to any successful AI strategy, but it's not always easy to do with AI investments. The benefits of AI can be both tangible and intangible, and they may take a while to become apparent. This complexity often creates a need for more clarity and confidence in AI investments, which can hold organisations back from fully embracing the potential of the technology. Demonstrating AI ROI allows you to justify the significant initial investments needed for AI initiatives, such as technology, talent, and infrastructure. By quantifying the value delivered by these investments, you can gain ongoing support from stakeholders and ensure the long-term success of your AI projects.
Furthermore, measuring ROI helps you prioritise initiatives based on their performance and potential impact. Not all AI projects are equal, and by comparing the ROI of different initiatives, you can allocate resources to those that provide the most value to your organisation. This data-driven approach to prioritisation ensures that you focus on the most impactful projects and maximise the return on your AI investments.
Measuring ROI for AI investments involves a structured approach that encompasses defining objectives, identifying metrics, tracking data, and calculating ROI.
Here's a step-by-step guide to help you effectively measure the ROI of your AI initiatives:
Define Clear Objectives
The first step in measuring ROI is to define clear, measurable objectives for your AI initiative. These objectives should align with your business strategy and address organisational challenges or opportunities. For example, your objective might be to reduce customer churn by 20% by implementing an AI-powered predictive analytics system. By setting specific, measurable goals, you establish a foundation for evaluating the success of your AI investment.
ROI Framework for AI Investments
Ensure you define SMART goals and break these into actionable use cases, prioritising based on ROI, implementation time, and complexity. Involving stakeholders from various departments ensures alignment and early buy-in for a successful AI implementation.
AI initiatives with a good market fit may cover any of the following:
Healthcare Diagnostics: enhances diagnostic accuracy and speed by analysing medical imagery.
Security: Provides real-time detection of security threats and malicious activities.
Document processing: automated the extraction and classification of information from documents.
Manufacturing: optimises manufacturing processes and predicts maintenance needs.
Inventory Management in Retail improves real-time tracking and predictive analytics for inventory management.
Agricultural Monitoring: aids in the timely detection of crop health issues and pest activities.
Autonomous Vehicles: interprets sensor data for real-time driving decisions in autonomous vehicles.
From a competitive advantage perspective, organisations that fail to embrace AI risk may fall behind competitors who leverage AI technologies to drive efficiency, productivity, new products, and increased customer satisfaction. This evaluation is more important than ever as Gen AI has opened the door for less AI-savvy companies to develop robust AI solutions quickly.
Identify Key Metrics
Identifying the key metrics that will help measure progress and determine your AI initiative's return on investment (ROI) is essential. These metrics should be directly linked to your goals and offer a comprehensive view of the project's performance. AI-related metrics may include cost savings, revenue growth, process efficiency, and customer satisfaction. Choosing metrics that are relevant, measurable, and in line with your business objectives is crucial.
Microsoft Research: Assessing AI Systems
Setting these metrics or Key Performance Indicators (KPIs) for AI investments is essential for monitoring the progress and success of your AI initiatives. Define KPIs that are directly associated with your strategic objectives. Assign responsibility for these KPIs to different departments and teams that can influence them. This accountability ensures that the insights obtained from AI models translate into actions. Ultimately, the actual value of AI lies in its ability to drive systemic change.
Data Plan: Track and Collect Data
Once you have set your objectives and metrics, it's vital to set up systems and processes to track and collect the necessary data throughout the AI project lifecycle. This may involve integrating AI tools with existing data sources, establishing data governance frameworks, and ensuring the accuracy and consistency of data collection. Regular data tracking is crucial for monitoring progress, identifying areas for improvement, and calculating ROI. Data quality is essential for successful AI implementations. Start by gathering all relevant data, whether within your organisational systems, hidden in various formats, or publicly available. Ensure that data is cleaned, transformed, and catalogued effectively. Remember, "garbage in, garbage out." Use POCs to identify systemic changes required to improve data quality continually. Minor process adjustments can lead to valuable new information. Consider professional advice or leveraging third-party training data operations software to help simplify core elements of the data process, including data labelling, model training, dataset management and customising workflows.
Calculate ROI
To calculate the ROI of implementing GenAI across the entire enterprise, it's essential to consider the costs involved and the expected benefits. This includes estimating cost savings, revenue increases, and other advantages AI can bring to the organisation. It's crucial to evaluate the financial implications of AI projects by using methods such as cost-benefit analysis, ROI calculations, and scenario modelling. Striking a balance between short-term gains, like cost savings and efficiency improvements, and long-term potential, such as revenue growth and market expansion, is essential. Additionally, it's important to measure the success of AI projects effectively by implementing robust performance metrics aligned with strategic objectives and KPIs. When developing AI models for your business, starting with a proof of concept is beneficial to validate your approach and ensure you have the right data, predictors, and analytical models to predict the desired outcomes. Once the proof of concept is successful, you can build and train your analytics model. Cost estimations should cover costs across software, hardware, cloud services, system software and talent.
Scaling/Measuring AI ROI for Long-Term Success
Artificial intelligence (AI) requires a solid understanding of strategic advantage, which means grasping how AI capabilities such as data infrastructure, teams, and skills can be enhanced. Without a connection to long-term advantage, many attempts at AI return on investment (ROI) result in what we refer to as "toy applications" – individual instances of AI that serve as experiments but do not contribute financial value or significant learning and skill to the investing company.
Furthermore, measuring ROI is crucial for building trust and securing stakeholder support, prioritising initiatives based on their impact, and fostering a culture of continuous improvement. By demonstrating AI's tangible benefits, organisations can gain the support and resources needed to expand their initiatives and maintain a leading position in an increasingly competitive market. Scaling AI is where the real benefits unfold.
Begin with modest investments in specific AI use cases. These smaller steps can generate substantial returns, potentially increasing revenue by up to 6%. As confidence and experience grow, AI investments ramp up. As AI initiatives expand, the revenue impact of AI can triple to 20% or more. Leading companies that scale AI outperform in EBIT and other Key Performance Indicators (KPIs), making scaling a critical part of your strategy.
Measuring ROI for AI investments requires a structured approach that involves defining clear objectives, identifying key metrics, tracking data, and calculating ROI. By following this framework, organisations can effectively quantify the value of their AI initiatives and make data-driven decisions to optimise performance and maximise returns.
