How to turn AI investments into measurable ROI: A CEO’s framework for creating business value

Business leaders across industries are eager to integrate AI more deeply into their organizations. But as more AI companies start adopting token-based billing, which has significantly increased the cost to use many AI tools, AI investments are becoming more of a financial risk.

How can you focus your AI spending on efforts that will deliver business value or a meaningful return on investment? Keep reading this AI ROI guide to find out.

What does AI ROI actually mean?

Proving AI value or establishing a clear AI ROI can be complicated. You might assume that you can achieve substantial cost savings in person-hours or realize new productivity gains by implementing an AI tool. But while this is sometimes the case, the actual value of an AI investment is often more nuanced.

Here are some specific areas where your business could explore an AI investment strategy to create ROI:

• Quality: In areas like coding, AI can help you produce higher-quality work. This is often the quickest way to achieve ROI, showing up faster than a potential uptick in productivity.
• Consistency: Likewise, AI can also help make your overall outputs more consistent, avoiding dips in quality or pace. This is also a relatively low-hanging fruit.
• Productivity: Rather than looking at AI as a tool to reduce person-hours, companies are expecting (and often getting) more productivity from many roles. However, meaningful productivity gains take time to achieve and may vary depending on the role.

Think in terms of solving specific problems or creating new opportunities

Never, ever start planning your AI investments by thinking about AI. That’s a recipe for aimlessly spending money with little to show for it.

Instead, identify specific problems or opportunities within your business — and only then start looking for AI tools to help you address them. This keeps you focused on activities that will deliver concrete business value.

Consider AI costs and ancillary benefits

Another key factor to consider here is the cost of a particular AI tool. Cost is a growing issue: After investing hundreds of billions of dollars in AI technology, companies like OpenAI, Anthropic and Microsoft are increasingly raising their prices and shifting to new billing models to close the significant gap between revenue and expenses.

Think carefully about how much it will cost your business to achieve gains in quality, consistency and/or productivity and whether you can adjust your usage to achieve meaningful results.

However, there are also ancillary financial benefits to being an AI-forward company that you may not have considered. Many investors want to know that you’re integrating AI into your core business, so doing so could help you raise additional capital or boost the value of your company.

And in certain cases, you can even claim federal R&D tax incentives for AI investments you make or deduct the full cost of AI expenditures upfront.

Why many AI initiatives fail to deliver measurable ROI

Despite AI’s potential to boost quality, consistency, productivity and more, many AI business investments fail to deliver measurable ROI. This is typically due to challenges or blockers like:

• Lack of focus: If you don’t focus your AI investments on solving specific business problems or developing clear business opportunities, you are likely to waste time, energy and money.
• Ineffective change management: It’s hard to roll out AI tools to people who don’t know how to use them, especially if your team is used to doing things a certain way. You need to approach AI implementation as a process of change — one that includes mindset shifts and training before you can expect to see results.
• Highly complex and fast-evolving technology: AI tech itself changes about every three and a half minutes. As a result, most organizations simply can’t keep up with the latest developments, let alone figure out how to make good use of them.
• No advisory support: Because of its inherent complexity, more businesses are turning to AI advisors for guidance on how to implement AI tools more effectively. This leaves organizations without such support at a major competitive disadvantage.
• Weak governance: Businesses with weak governance policies risk problems like shadow AI, in which team members use unauthorized AI tools with unpredictable consequences.
• Poor data readiness: The saying garbage in, garbage out, holds true for AI. If you don’t have clean, organized data to feed into your AI models, you won’t generate useful outputs.

A framework for turning AI investments into business value

Developing an effective implementation framework can help you turn your AI investments into business value or ROI. Here’s a basic framework you can use as a jumping-off point:

1. Start with policy and governance

Your governance policies set the tone for how you use AI in your business. Effective governance can both deliver better results and reduce your risks, so make this your start point before tackling anything else.

2. Find an AI advisory partner

Because AI technology is evolving so quickly, your team almost certainly doesn’t have the bandwidth to stay current on the latest developments. An advisor can help you understand where AI is now, where it’s going and how you can use it to strengthen your business.

3. Identify specific problems, opportunities, use cases and people

Spending money on AI won’t do much for you unless you’re focusing your efforts on specific problems, opportunities and use cases. Also important: Identify people within your organization who are excited about AI and can serve as early-adopter champions, whose enthusiasm will help spread AI use among their peers.

4. Think about how to monitor AI use and control costs

Consider how you’re going to monitor AI use within your organization and control costs. As more AI providers move to token-based billing, this will become even more essential (if you don’t want to get hit with a shocking bill). You’ll want to ensure you have controls and risk management policies in place to both understand how your organization uses AI and keep your spending right-sized.

How CEOs should measure AI ROI

How should you go about measuring your AI success or ROI? Business leaders have often focused on usage or licensing rates, but these are largely vanity metrics (Amazon workers famously gamed the company’s AI usage leaderboard by assigning their AI agents to perform meaningless tasks). Meanwhile, the reality that different roles use AI differently also makes comparing usage rates an apples-to-oranges exercise.

Instead, focus on metrics that show you whether AI is making a measurable impact on your business results. The key here is to treat this effort like an experiment. Are you seeing changes in productivity, quality, customer satisfaction and other KPIs? Is there a meaningful difference in these areas between employees who use AI and those who don’t?

If your answer to some of these questions is yes, you’re probably seeing ROI.

Where AI is delivering value today

You can use AI to create value in multiple aspects of your business. Key areas to explore include finance, operations, HR, sales and customer experience.

Finance

AI can help your finance team speed up or fully automate many repetitive accounting and bookkeeping tasks. For example, you can deploy AI to tackle accounts payable tasks like processing invoice images and comparing the data against what’s in your system and can also use AI tools to assist your team with financial modeling and reporting.

Operations

Your operations team can use AI to automate routine tasks like putting leads into your CRM, data entry, coordinating meetings or generating meeting summaries and moving information from point A to point B. AI can also help you review regulations, understand if you are in compliance with oversight rules and document your compliance efforts.

HR

HR teams are increasingly exploring AI as a tool to assist with the hiring process. For example, you could ask an AI tool to compare a job candidate’s resume with the posted job description and then generate a list of questions to ask during their interview. However, HR is also a good example of why you need a strong AI governance policy in place, as you could run afoul of security and privacy regulations here if you don’t keep your AI use in line with standards.

Sales and customer experience

Sales teams are using AI to research prospects or potential clients, while their customer experience counterparts are turning to AI to help keep up with existing clients. For example, you can set an AI agent to search for news about a key client or prospect, or to pull insights from a database like ZoomInfo.

You can also use AI to quickly review thousands of your own recorded sales or customer calls to create scripts and follow-up questions based on what has proven effective with your specific prospect or customer base.

Moving from AI pilots to enterprise-wide adoption

You can launch an AI pilot with relatively little effort. However, notable ROI usually comes from scaling that pilot up to enterprise-wide AI adoption after proving its value.

This process demands a deeper level of change. As a basic framework for scaling your AI use to maturity, picture a four-layer cake:

• Layer 1: Data. Build a clean, organized data foundation for your AI tools to draw on.
• Layer 2: Execution. Create AI agents that can understand and use your data to complete basic tasks.
• Layer 3: Intelligence. Add in domain-level agents that serve as the AI agent equivalent of department heads that can provide context on your business strategies, goals and daily operations to your lower-level AI agents.
• Layer 4: Conversation. Finally, you want to layer a communication agent on top of everything else, one that understands your overall business and can both direct your other agents and make strategy recommendations to human users.

FAQs about AI ROI

Here are commonly asked questions about AI ROI:

How long does it take to see ROI from AI?

Different tools have different timelines. With an AI assistant, for example, you want to see production improvement within 4-6 months. If you don’t, it’s not being implemented correctly, it’s not the right tool, or it’s been given to the wrong people.

However, bigger investments may take longer. While brand-new companies are starting with enterprise-level AI in mind, existing companies often need to make a mindset change before seeing an AI ROI.

Mindset changes lead to process changes that result in outcome changes, so focus on mindset before investing heavily in enterprise AI. You won’t pay off enterprise-level investments quickly, but you should start seeing some gains right away.

What is a good ROI for an AI investment?

There is not necessarily one number to aim for here. Instead, cutting-edge companies are looking at AI expenses as a percentage of payroll — essentially treating the money you’re investing in AI credits for your employees as an employee-related expense like a uniform or a piece of equipment.

The percentage of payroll that AI should make up will vary by industry and will be different for different roles. Look for benchmarks based on your specific industry to get a clearer sense of what a reasonable percentage might look like for your business.

Which AI initiatives typically generate the fastest ROI?

If you’re looking for low-hanging fruit that can quickly deliver ROI on an AI investment, consider knowledge agents or AI assistants. These are fast to implement and can help your team by making it easy to access organizational information, get answers to common questions and unblock small barriers to productivity.

What are your next steps toward better AI business outcomes?

To improve your ROI on your AI investments, find an advisory firm that offers AI consulting services. A good advisor will know your specific industry, understand the latest developments in AI technology and help guide your business forward with an eye toward how AI continues to evolve.

How Wipfli can help

We advise businesses on using AI to solve organizational problems and drive growth. Let’s talk about your goals and how a more effective AI strategy can help you reach them. Start a conversation.

Let’s make AI work for your business

Read more

• As AI spending grows, companies must rethink capital planning
• AI governance checklist: A basic framework for your business
• Salesforce failed to replace thousands of workers with AI. What can your business learn?