Beyond the Hype: A Problem-First AI Strategy

AI is the future. At least that is what many companies today base their growth on. However, while everyone wants to or is already jumping on the AI train, what may come as a surprise is that almost no one is profiting from it.

Yes, you’ve read that correctly.

Different data portray a surprising chasm between AI investment and the return on that investment. AI is the future, but it cannot impact ROI or overall company growth if it is done solely for the sake of FOMO: ‘our competitors are doing it, so we must too.’

There needs to be a clear strategy and a defined business case. What problem is AI solving for your company? How is it shaping user experience and satisfaction, and how is it influencing ROI and overall growth? If you can’t measure the success of your AI initiatives, it becomes impossible to know whether they’re simply adding to your costs or genuinely driving business performance and strengthening trust in your brand.

The AI Hype Bubble and Zero ROI

In 2024, 78% of organisations reported using AI in their business, which presents a major leap from 55% in 2023. Use of generative AI in at least one business function has more than doubled, from 33% to 71% in the last year alone.

MIT’s research project titled ‘The Gen AI Divide State of AI in Business 2025’ analysed over 300 businesses and their deployments of generative AI. This research uncovered that 95% of these companies had zero return from their investments in generative AI.

The problem is not in the quality or the possibilities that generative AI offers. It’s in the approach. In other words, the 95% failure rate is a direct result of the “tech-first” approach caused by the AI hype rather than a “strategy-first” or a “problem-first” approach.

The successful 5% minority start their projects with the problem they need to solve, creating a clear, visible, and measurable AI strategy. By ensuring that all the initiatives are directly tied to measurable business outcomes, this approach doubles their chances of experiencing revenue growth.

So, start with a question like this one: “What are our most expensive, unsolved business problems, and how can technology provide a solution?” This way, the potential AI investment is directly connected to your business priorities from the very beginning.

For marketers, this means identifying the precise points where revenue is leaking and you are losing the trust of your customers. It also ensures that the different departments work together in solving the problem, not leaving it to the IT department that may not be familiar with the root problem or what needs to be done to solve it.

Let’s apply this "problem-first" framework to the two biggest, most expensive leaks in your marketing funnel right now.

Problem #1: The $260B Wound You Can't Stop

Let's start with the problem every marketer tracks: cart abandonment. You already know the 70% average rate by heart. But many teams have come to accept this as "window shopping."

The data shows this is wrong. This is not passive browsing, but active frustration. The top reasons for abandonment are unexpected shipping costs (48%) and a complex checkout process (18%). These are your high-intent buyers hitting a wall of friction, and it’s a $260 billion recoverable revenue opportunity.

For the last decade, the solution has been reactive. Letting the customer get frustrated, leave, and then try to win them back 24 hours later with a cart recovery email.

The "problem-first" AI solution is proactive Conversational AI. Instead of recovering the sale after you've frustrated the customer, you prevent the abandonment from ever happening.

This solution uses an AI assistant to intervene at the precise moment of friction. When a customer hesitates on the payment page, the AI doesn't wait. It proactively clarifies the shipping policy or even offers a discount to solve the problem in real-time.

The measurable ROI for this approach is staggering. A 2024 Forrester Total Economic Impact (TEI) study on this exact strategy found it delivered a 249% Return on Investment. It helps recover 35% of abandoned carts in-session and drives a 4x higher conversion rate (12.3% vs. 3.1%) for users who engage.

Problem #2: The $2 Trillion "Silent Killer" You're Not Measuring

While cart abandonment is the leak you see, there's a much larger, "silent killer" you are likely miscategorising: search abandonment.

This is the customer who uses your site search, gets a bad result, and immediately leaves. You probably classify this as a generic "bounce," but it's a high-intent buyer you've just failed. This single failure point costs retailers over $2 trillion globally each year.

This is where we must address the "advanced search" question. The $2 trillion problem is caused by legacy, keyword-based search. This is the technology on most e-commerce sites today. It's a rigid tool that can only find exact keyword matches.

So, when a customer types a natural, human question like, "What should I wear for jogging in cold weather?", or even misses to type the exact keyword, the legacy system fails. It looks for those keywords in product titles, finds no match, and returns "0 results." The customer is gone.

The "problem-first" solution is the AI-powered search. It does not match results based on keywords, but based on customer’s intent.

When a customer asks that same question, the AI understands the intent behind the words. It knows the query means "winter running jackets," "thermal-lined tights," and "base layers," and it returns those relevant products.

This is the AI solution that directly fixes the $2 trillion leak. The measurable ROI is immediate. Data shows that AI-powered search delivers 2-4x higher conversion rates than traditional search. Even more powerfully, it expands the sale. After a successful AI-powered search, shoppers purchase, on average, three additional items besides the one they were looking for. You've transformed your failed search bar into your store's most effective AI salesperson.

Conclusion

The 95% of companies failing at AI are asking: "What can we do with AI?"

The 5% who are winning are asking, "What problems can we solve with AI?"

Your path to AI-driven growth doesn't require a massive initiative. It requires focused projects related directly to your biggest and most expensive business issues. That is how you de-risk your investment and build an AI strategy that delivers measurable, bankable ROI.