One of the most iconic lines from Tom Cruise’s movie “Jerry Maguire” is him screaming “show me the money” over the phone to one of the athletes his character represents. A new survey by McKinsey & Company shows that companies that’ve deployed generative artificial intelligence (GenAI) may be asking the same thing. Where GenAI companies will be raking in $1.1T in global revenue by 2028, up from $45B in 2024, according to a report from Morgan Stanley, the businesses using the technology are not reaping much financially from the investment.If your bank is frustrated by the generative AI hype but lack of bottom-line improvement, you’re not alone. The McKinsey survey found that 8 out of 10 companies that deployed GenAI reported “no significant bottom-line impact.” McKinsey’s conclusion may be conservative. An MIT report found that organizations have invested between $30B and $40B in GenAI, but 95% realized no measurable return at all. But don’t give up on it just yet. The new technology still shows promise of major financial benefits in the future.The GenAI ParadoxThe explosive growth of artificial Intelligence is the latest tech innovation to storm the business world, with nearly 8 out of 10 businesses reporting that they have used GenAI. The disconnect between sky-high adoption and lack of bottom-line improvement is what McKinsey called the “GenAI paradox.”In fact, GenAI may simply be undergoing a return-on-investment arc similar to earlier tech innovations, with slow-to-emerge gains in productivity. Early adopters often stumble, integration into the organization faces obstacles, and rewards are hard to find — at least initially.The Deployment ProblemOne key problem, according to MIT, is how GenAI is being deployed. Often, the technology takes the form of general-use chatbots or employee copilots that are hard to track and quantify. More focused, specific uses can be more effective and quantifiable, but have proved difficult to deploy successfully. Custom-designed GenAI systems rarely make it out of tests or trials.So, what is a community financial institution to make of all this at a time when GenAI solutions are being hawked with ferocious intensity?Four GenAI Strategies To Help Increase ROIDespite being a prolific tool, GenAI isn’t something you can invest in haphazardly and still expect profitable results. The best GenAI plans are careful and intentional. Here are four strategies that can help you find more bottom-line success with GenAI:
- Stay focused on customers. In the past, tech innovations that panned out helped organizations serve customers better, faster, and cheaper. GenAI will likely do the same, given time. When looking at a GenAI solution, ask yourself this question: how will this tool benefit my customers or how I serve them?
- Concentrate on back-office functions, which deliver the fastest ROI. About half of GenAI spending has been going to sales and marketing, which are departments that are slow to show bottom-line results for this type of tool. Outsourced GenAI back-office solutions actually reduced overall business process outsourcing costs by $2MM to $10MM annually.
- Don’t try to go it alone. Companies that tried to use internal resources to develop and deploy GenAI were far more likely to fail than those that sought outside help. Companies that developed partnerships with outside experts and bought systems from established vendors successfully deployed 67% of the time, compared to 33% for custom internal builds.
- Prioritize the best move rather than the fastest move, particularly if your tech expertise isn’t completely ready for GenAI. While it may seem like you are involved in a sprint to deploy GenAI, going too fast can lead to costly mistakes. The number of organizations that abandoned most of their AI initiatives before broad adoption soared to 42% this year, versus 17% last year. Conclusion: a slow, deliberate approach can be very useful at this moment.
While today’s numbers may seem discouraging, community financial institutions should remember that every major technological breakthrough follows a similar path — initial hype, early setbacks, and then a steady climb toward real value. Generative AI is no different, and institutions that take a thoughtful, customer-centered, and disciplined approach will be well-positioned to capture its eventual rewards. The institutions that treat this period as groundwork — experimenting carefully, building expertise, and preparing the right foundations — will be the ones that see outsized gains when the technology matures into a true driver of efficiency, growth, and customer loyalty.
