The initial excitement for generative AI (GenAI) has cooled. And now many brands that invested in early experiments are asking a fair question: Where’s the business value for generative AI in ecommerce?
This shift is so common that the technology research firm Gartner calls it the “Trough of Disillusionment.” It’s a natural phase where excitement gives way to the practical challenge of making a new technology deliver on its promises.
The data shows why. A recent MIT report found that 95% of organizations are getting zero return from their GenAI investments.
But this isn't a story about failure. It’s about a necessary shift toward a more mature approach. While most companies are struggling, a small group — just 5% — are achieving significant results by using AI to solve real business problems. They have a clear strategy that any brand can follow.
This moment is an opportunity to build a real advantage by focusing on what actually works.
Why Many AI Projects Fail To Deliver Value
The gap between companies that succeed with AI and those that don’t comes down to approach. The reasons are consistent and have little to do with the technology itself. Most stalled initiatives share the same common problems.
Focusing on Technology, Not Business Problems
Many early projects were driven by a fear of missing out, leading companies to adopt AI simply to appear innovative. They took a "checkbox" approach rather than identifying a specific business problem to solve, according to Gartner.
Creating Disruption to Workflows
The top reason employees reject new AI tools is that they don’t fit into their daily work. The tools are often difficult to use, don’t learn from feedback, and aren’t well-integrated with existing systems, according to MIT.
Investing in the Wrong Areas
Companies often direct their AI budgets toward visible functions like sales and marketing. However, MIT’s research shows that the clearest and most significant financial returns are often found elsewhere in the business.
A Practical Guide to Generative AI Ecommerce Success
For brands that want to get real value from AI, the path forward requires a mindset shift. The goal isn’t just to use AI, but to enable the business with it.
As Rob Gonzalez, Salsify co-founder and chief strategy and innovation officer, writes in his whitepaper “The PXM AI Manifesto,” the most powerful use of AI is not in replacing people, but in accelerating them.
Build Your Strategy on Measurable Business Results
An AI project shouldn’t be an experiment; it should be a solution to a problem. Every initiative must be tied to clear business value and a measurable return on investment.
Interestingly, the biggest returns are often hidden in plain sight. While many companies focus AI on sales, MIT found that the most dramatic cost savings come from back-office automation.
For example, MIT reports that some companies have seen:
- $2 million-$10 million in annual savings by automating tasks previously handled by business process outsourcing firms.
- A 30% reduction in spending on external creative and content agencies.
These are real cost reductions that directly improve financial performance. By focusing on the right goals, you can develop a compelling business case for an effective AI strategy. To help, you can start by identifying the ecommerce key performance indicators you should track.
Make Your Product Data the AI Source of Truth
AI systems learn from the information they can access, so if your product data is inconsistent, incomplete, or inaccurate, your content likely won’t surface. A strong data foundation is no longer optional; it’s the foundational fuel for the AI-powered digital shelf, as well as the rapidly developing era of agentic commerce.
Gonzalez says the first principle of success is to "make your product data the AI source of truth." This is because your product content — from your website and product pages to customer reviews — is the training material that shapes how AI represents your brand.
Think of it this way: Your content now serves a third audience beyond customers and search algorithms. It must also be clear and structured for AI agents.
Choose Partners and Tools That Fit Your Business
AI success rarely comes from building a tool from scratch. Research shows that projects developed through strategic partnerships are twice as likely to succeed as those built internally.
The key is to find technology that fits your business, not the other way around.
According to MIT, when selecting GenAI vendors, business leaders are clear about what they want:
- Nearly 75% of executives demand a deep understanding of their workflows
- About 70% of executives require minimal disruption to their current tools
The best AI solutions act as a "copilot," not an autopilot. They empower your team to work more effectively within their existing processes. This approach also ensures you maintain control.
As Gonzalez says, "scale without control is chaos," and the right systems provide the governance to keep your team in charge of brand integrity.
How Generative AI in Retail Works in Practice: The Ubique Group Example
Home goods retailer Ubique Group needed to manage a massive and complex product catalog for brands like Humble Crew and Costway. Creating and optimizing content for more than 20,000 products across dozens of retail channels was a significant operational challenge.
Focus on What Works To Move Forward
The "trough of disillusionment" is actually a positive development. It forces a hype filter on vendors and clarifies what it takes to succeed with AI.
While other companies are dealing with stalled projects, your brand can gain a significant advantage. The roadmap is straightforward: Focus on measurable results, establish your product data as the source of truth, and choose partners and tools that fit the way your teams work.
This approach allows AI to handle the volume and complexity of the digital shelf, freeing up your team to do what they do best: Lead the vision.