By: Natalie Johnson
Another explanation of how AI works is unlikely to change the minds of skeptics. What can make an impact is a clear connection between the technology and the business problem sitting in front of them. For Aicha Gersing, Senior Vice President of Customer Success for Agentforce and Data Cloud at Salesforce, that distinction is central to making AI conversations meaningful.
“The first thing that one has to do is stop explaining AI and start explaining the before and after of a specific workflow they’ve already cared about,” Gersing says. Executives tend to respond to evidence that can be measured in operational and financial terms. A team completing a process 400 times a month, for example, provides a concrete baseline for evaluating what AI changes. The conversation can then move from abstract capabilities to costs, time, productivity, and revenue.
“Skeptics don’t need AI competency memos. They really need a before and after that holds up to their scrutiny,” Gersing says, stressing the importance of resisting the temptation to lead with what AI can theoretically do. Instead, the strongest cases show how a specific workflow performs before and after implementation.
Technical support is one place Gersing sees this approach working well. Routine support interactions follow repeatable patterns, so the before and after can be tracked in resolution times and team capacity rather than in broad claims about what the technology can do.
Data Anxiety Can Mask the Real Objection
For many executives, skepticism about AI is closely tied to uncertainty about their data. The conventional message that successful AI requires high-quality, well-architected data can unintentionally reinforce hesitation. Leaders may conclude that their organization is simply not ready. Gersing sees a more complicated concern underneath that response: executives may not want to discover the weaknesses in their data infrastructure through a highly visible AI pilot.
The answer is not necessarily to wait until every data challenge has been solved. A more practical approach is to start with a focused use case where sufficient data already exists and use it to test what is possible. Increasingly, tools that aggregate and contextualize information can also reduce the importance of where data physically sits.
Conviction Shows Up in the Questions
The shift from skepticism to conviction is visible in how executives talk about AI. Gersing identifies three signals:
• First, skeptics begin defending the technology to others. The question changes from “show me” to “let me explain this to my chief financial officer (CFO).”
• Second, they start asking about ownership, structure and resources.
• Third, they stop asking for proof and start asking about speed. “Skeptics ask ‘Does this work?’ Believers ask ‘How do I scale it?’” Gersing says.
That distinction matters because adoption comes down to moving the organization from experimentation toward a repeatable operating model.
The AI Leaders Close the Loop
The companies generating both revenue growth and cost savings from AI are not necessarily the ones running the most pilots. They are the ones learning from deployment. Gersing describes these organizations as operating with feedback loops. They measure performance, iterate and improve, rather than treating an AI implementation as a one-time project. “They treat it as a compounding system rather than a project,” she says.
Internal productivity can create meaningful gains, but customer-facing applications can affect both sides of the financial equation. Support deflection, onboarding, and risk identification can reduce costs, while improving the customer experience and creating opportunities for growth. The implication is significant. AI maturity may be less about having the most sophisticated technology than about building the organizational discipline to learn from it.
From AI Experiment to Boardroom Language
As AI adoption matures, some decisions that currently appear advanced are likely to become standard business considerations. Gersing expects choosing between deterministic automation, reasoning agents, and multi-agent systems to become a routine architectural decision, much like choosing between a spreadsheet and a database.
“Matching architecture to workflow will be table stakes vocabulary in boardrooms,” she says. That shift will also change how executives discuss AI investment. Gersing anticipates that questions about AI spending as a percentage of revenue will increasingly become part of CFO-level expectations rather than purely strategic technology discussions.
For skeptical executives, then, the path to AI conviction is evidence, iteration, and business relevance. The most persuasive AI story is ultimately the one that makes its impact visible in the numbers, creates a mechanism for continuous improvement, and gives leaders a credible answer to the question that matters most: “What should the organization do next?”
Follow Aicha Gersing on LinkedIn.
