With ChatGPT’s user base setting growth records and NVIDIA’s market cap reaching over $3 trillion, it’s clear we’re in the midst of a major AI boom. Organizations across industries are rushing to invest in AI tools that claim to automate and optimize business activities from email copywriting to enterprise data governance—and cut costs to boot.
But how will companies look back on these initial AI investments in five years? In 10? If this hype cycle is anything like the ones that came before it, much of today’s hurried AI spending will never show significant ROI.
It’s not that AI isn’t a powerful tool or that it doesn’t have the potential to transform business operations for the better. But too often during boom periods like these, companies rush to adopt new technologies without taking the time to deploy them thoughtfully and efficiently.
As a result, new solutions never deliver the payoff that was promised. For example, four years after the first COVID-19 lockdowns, many organizations are still trying to realize ROI from digital investments made during their pandemic-era tech spending sprees. Capabilities like mobile ordering and touchless payments are great investments, but if you build them on the wrong platform, you might struggle to scale them as your business evolves.
The good news is that today’s organizations don’t have to make the same mistakes with AI. A thoughtful approach focused on maximizing ROI from digital investments can help you lay the groundwork for reaping the full benefits of this revolutionary technology in the future.
THE PITFALLS OF INVESTING IN NEW TECHNOLOGY TOO FAST
The rush to invest in AI is a common pattern with new technologies, from ERP systems in the 1990s to cloud in the 2010s and touchless ordering during the pandemic. In each of these cycles, companies that jump in too fast, too early tend to encounter many of the same problems, including:
• Lack of supporting infrastructure
When an organization adopts a new solution too quickly, they often realize later that they’re missing the necessary infrastructure for the solution to deliver its full functionality. For example, many companies rushed to set up new digital e-commerce solutions in response to pandemic-driven changes in customer behavior.
Some did so without thinking through how they would structure and manage the large quantities of data their digital platforms collect. Thanks to internal silos and other issues, they struggle to leverage that data for activities that would generate bottom-line results.
• Difficulty keeping pace with technological growth
Innovative new technologies evolve quickly, which means the solutions you build on or with must evolve quickly too. For example, most apps designed to run on a 4- or 8-GB first-generation iPhone would look and feel outdated running on a 16- or 32-GB iPhone 4 just three years later.
If new technology is a critical element of an application or product, you should expect to rebuild, retrofit, or update that product on a regular basis to take advantage of new features and capabilities. Organizations that don’t account for these necessary adaptations in their IT roadmaps often find themselves playing catch-up as underlying technologies evolve.
• Difficulty evolving the customer experience
Keeping up with customer expectations demands that you think ahead to predict future shifts. For example, right now the AI assistant Microsoft Copilot is available as a standalone app—but what will mobile experiences look like when Copilot is natively embedded on every app on your phone? Organizations must anticipate these types of changes by continuously upgrading their user experiences or they’ll fall behind more agile competitors.
HOW TO MAKE THE MOST OUT OF YOUR AI INVESTMENT
It’s never too late to fix these issues and realize ROI from rushed or ill-considered digital investments. However, by proceeding methodically and adopting new technologies thoughtfully, you can avoid needing to make painful adjustments later.
For AI specifically, there are a few key steps to consider:
1. Get your data house in order.
AI is only as good as the data it’s trained on. So now is an excellent time to perform some basic data hygiene if you haven’t already. Work to break down internal silos, centralize data management, and establish clear governance guidelines. This will also improve your analytics and any other non-AI data-driven functions.
2. Identify impactful use cases.
It’s relatively easy for any company to deploy a customer service chatbot or an AI assistant to help the marketing team write better emails. But that’s likely not where AI is going to make the biggest impact. Instead, search for use cases that touch your organization’s core business. Then, consider whether existing AI tools can help make them faster, cheaper, more accurate—or all three.
3. Prototype, prototype, prototype.
Don’t jump right into rolling out a new AI solution across your entire organization. You’ll need to prototype and test your solution, whether it’s built in-house or out of the box. Ideally, you want more than just a quick pilot program.
Depending on the solution’s sophistication, multiple rounds of iteration might be needed to get things right. Also, bear in mind that the iterative process doesn’t end with the launch. Your solutions will need to evolve alongside the technologies they’re built on.
4. Lean on partners.
The learning curve with AI is steep, and hiring talent with experience in the field is expensive. An outside partner with experience guiding multiple organizations through AI deployments can bring much-needed insight to your process (full disclosure: Bridgenext provides this service). Besides technical expertise, look for a partner with cross-functional experience who can offer advice on best practices across the industry.
AI REQUIRES A THOUGHTFUL APPROACH
AI solutions have the potential to transform how we work and accelerate organizations’ progress toward their most important business goals. But as with any digital investment, you must deploy AI thoughtfully or you won’t realize a return.
Before you make the leap into a new AI initiative, take the time to identify use cases, prototype, and test—with the help of an experienced partner if possible. Five or 10 years down the line, your bottom line will reap the benefit.