Retailers should cut through the AI hype for tangible results.
By Karthik Ganapathi
The conversation around artificial intelligence is impossible to escape. For executives, it can feel like standing in the middle of a crowded street: There is a lot of noise, but not much clarity. But for business operators, the path forward doesn’t lie in speculation or hype—it lies in results. AI is not a magic wand, nor is it a looming threat waiting to replace the workforce. It’s a tool. Like every other industrial innovation, its value comes from how it’s applied. We already experience AI every day: Netflix recommendations, spell check, sentence completion prompts, voice assistants, fraud protection and assisted driving all augment human decision-making without taking control away from humans.
Before applying AI, it’s important to clarify what we mean by it. One helpful distinction is narrow versus general AI. Narrow AI builds targeted intelligence into systems to help make faster, better decisions and is already delivering value in industrial automation, analytics and customer operations. General AI, the so-called holy grail, would pass the “Turing test,” having the ability to understand, reason, learn and behave at a human level. However, it remains largely hypothetical and the source of sweeping claims—often driven by marketing buzz.
Too often, companies either hesitate to act for fear of making the wrong move or throw themselves into pilot programs that never leave the lab. The reality is that successful AI adoption requires something much simpler: focus. Start with a business challenge that matters—reducing downtime, improving quality, streamlining logistics. Apply AI to those challenges, measure its impact and scale what works. Then translate the theoretical promise to tangible outcomes: move from “can be done” to “we did it.”
The truth is that the hype around AI will fade—what won’t fade are the competitive advantages for the companies that take a disciplined, practical approach to AI. For executives looking to cut through the hype, three principles provide a clear starting point:
- Get your data right. AI is only as good as the data you feed it. Without clean, structured and harmonized information, even the most advanced models will generate noise or misleading results. Start by organizing customer, product and equipment data into a single source of truth. This approach provides visibility into profitability, operational performance and opportunities for improvement, creating a foundation upon which any AI-enabled system can reliably operate. This view also helps organizations analyze current performance, identify gaps and prepare for more advanced AI applications.
- Start small, scale fast. The most effective AI implementations begin with targeted, high-friction problems rather than broad, undefined initiatives. At Vontier, for example, a spike in call volumes prompted the team to apply the old-school Kaizen methodology, identifying areas for continuous improvement across the organization. This analysis clarified which application and process changes were needed to reduce call volumes. With that foundation in place, the company leveraged agentic AI tools to process incoming calls more efficiently. Some improvements were simple, such as deploying AI agents to augment support personnel. Others—like automated triaging and closed-looped backlog management—required more advanced AI models and careful orchestration. This first step demonstrated that disciplined, small-scale pilots can deliver tangible results and build a foundation for broader adoption.
- Keep humans in the loop. AI can predict outcomes, generate insights and automate tasks, but it cannot replace human judgment. Leaders should treat AI as an assistant rather than a substitute, ensuring outputs are validated, errors caught and biases mitigated. Human oversight safeguards quality and accountability while allowing teams to redirect effort toward higher-value work—interpreting insights, making strategic decisions and innovating in ways machines cannot replicate.
Practical applications for AI span multiple domains: In product design, software development and delivery, AI can provide near-immediate benefits, with predicted efficiency savings of $2.6 to $4.4 trillion. AI accelerates discovery, isolates core feature needs and translates them into discrete engineering tasks, reducing cycle times and improving roadmap fidelity.
The bottom line is clear: AI today will not fully replace operations, but businesses that embrace AI with discipline and purpose will outpace those who don’t. The winners won’t be the ones who talk the loudest about AI—they’ll be the ones who get their data right, solve real problems and keep humans at the center of decision-making.
Karthik Ganapathi is the president of Invenco by GVRInvenco by GVR, a Vontier business. Learn more at invenco.com.