The Roadmap to AI-Driven Sales Teams w/Craig Nelson

Sales teams do not need another vague promise about AI transforming everything. They need a practical way to move from curiosity to a repeatable process that saves time and improves the work around each customer conversation. That was the focus of my conversation with Craig Nelson, a sales enablement leader who was helping teams introduce AI while scaling with fewer resources.

Craig had seen organizations add products, locations, and people while still relying on manual habits. His starting point was simple: look at the work required to prepare for a sales call, run the call, and debrief afterward. Those steps contain opportunities for AI, but the technology only matters when it is attached to a clear process.

Begin with the work around the call

The fear that AI will take someone's job is usually the first issue to address. Craig put it directly: “AI is not going to take your job but somebody who knows AI just may.” The point is not to scare a team into adoption. It is to show people that learning a useful workflow can give them time back and make their expertise more valuable.

Sales calls are an obvious place to start. AI can help prepare a rep, capture what happened during the conversation, and create the follow-up afterward. A transcript can be checked against the topics that needed to be covered. At the end of a call, the system can surface what was understood, identify action items, and suggest what should be addressed in the next meeting.

That kind of workflow can also become a CustomGPT. If every call follows a similar structure, give the system instructions for what to find in the transcript and how to format the email or action list. The value comes from connecting several small tasks, rather than expecting one tool to solve sales enablement by itself.

Prove value with one useful workflow

Craig recommended finding a practical starting point instead of asking everyone to become an AI expert at once. A customer-facing role is a strong candidate because the time saved can affect revenue or retention. Pick a repeated task, document the current process, test the AI-assisted version, and measure the time or quality change.

The goal is to show a real gain quickly. Craig talked about saving 60 to 80 hours in the first month per person as a way to justify the software cost. That calculation also forces a team to look for overlapping tools and unnecessary subscriptions. Technology should simplify the work, not create a second layer of administration.

Once the workflow works, repeat it with another person and then another team. That is how an organization moves from isolated experiments to an institutional process. Training should include the reason for the workflow, the steps to follow, and the points where a human needs to review the output.

Discipline makes the difference

The long-term roadmap is not a collection of random AI experiments. It is a sequence of proven use cases that returns time to people, improves coaching, and creates better feedback loops. Only after a process has been tested should it be rolled out more broadly.

The same discipline applies to choosing what to automate. Start by identifying the greatest pain point, especially in work that makes or keeps money for the company. Then build the smallest useful workflow around that problem and learn from the results.

AI can support sales teams before, during, and after calls. The advantage comes from deciding where it belongs, measuring the result, and giving people a process they can use consistently.