Where AI Belongs In Your Sales Process (And Where it Doesn’t)

You need to know what’s being used and how.

Before you lay down rules, you have to understand the landscape. That means asking a few direct questions:

• What AI tools are your reps using?

• What parts of the sales process are they applying those tools to?

• Are those tools feeding into emails, decks, call summaries, or CRMs?

• Is anything AI-written going directly to the customer without a second set of eyes?

This isn’t a crackdown. You’re not there to scold. You’re there to understand the lay of the land so you can lead with clarity instead of assumption.

When you bring these conversations into the open and your reps start to talk and collaborate naturally, you’ll start to see the usage patterns emerge. One rep might have a killer AI set of prompts for turning lead data into simple bullet points to work into your outreach. One might be using it to help keep specific customer needs top-of-mind.

The opportunities are endless; you’ve just got to bring them out into the open.

Your job is to set expectations, not stifle momentum.

Once you know what tools are in play, and start getting some rough ideas on where there’s some wins to seize on, you can start outlining what’s acceptable and what isn’t. But that only works if the expectations are simple and repeatable. Here’s where I’d start:

• Be transparent. If you’re using AI in any part of your workflow, say so. Make it normal to talk about the tools being used and the outcomes they’re helping create.

• No misrepresentation. AI should never be used to make promises the company can’t keep, position the product inaccurately, or shift tone in a way that goes against your brand.

• Stick to the facts. AI can help with efficiency, but it’s your rep’s responsibility to ensure accuracy and clarity in everything the customer sees.

If you put those expectations in plain language, your team will start to self-correct. You don’t need a 30-page policy. You just need shared understanding and clear expectations.

Let your team find a shared rhythm.

When you let reps talk openly about how they’re using AI, something interesting happens. They start to find a common water level.

You’ll see the more experienced reps share what’s working. You’ll see the ones who are skeptical begin to try a few safe use cases. You’ll see language start to align and best practices begin to form from the ground-level and bubble up.

That’s how teams build smart, effective habits. Not from a rulebook, but from shared wins and honest reflection. And it only happens if you give them the space to talk about it.

Set aside an hour a week. Ask three simple questions:

• How did you use AI this week?

• Where did it help?

• Where did it fall flat?

Let the team talk. You’ll be surprised how fast clarity starts to emerge.

Organizing call notes and CRM data

If you’re managing more than a handful of reps, you’ve probably seen the same thing again and again: great conversations followed by terrible documentation. Details get lost, notes are vague, and nobody wants to go back and sort it all out.

AI can clean up that mess. When integrated with your CRM or used to summarize transcripts, it helps keep your team focused on the next step instead of getting stuck in admin work. It won’t do the thinking for them, but it will make it easier to keep track of what matters.

If your team isn’t using AI for this already, it’s one of the fastest ways to give them back time without taking away control.

Helping draft first-touch cold outreach

Cold outreach still matters, but it’s often one of the most mentally draining parts of the job. You want reps reaching out with confidence, not second-guessing every word of a first message. That’s where AI can help generate a working draft, but not a final version.

Let me be clear: this doesn’t replace the research. Your team still needs to understand the prospect, the industry, and the reason they’re reaching out. But if AI can help them move past the blinking cursor and start with something usable, that’s a win.

They should still be editing for tone, accuracy, personalization, and context. But using AI to help kickstart the writing process is absolutely fair game if they’re doing the homework first.

Rephrasing or refining internal messaging

Reps aren’t just writing to prospects. They’re also writing to you, their managers, to support teams, or to each other. Sometimes they struggle to find the right way to phrase something or explain a need clearly.

AI can step in here as a second brain. It can help reword clunky language, turn bullet points into a structured update, or clean up a follow-up note without having to start from scratch.

This isn’t about customer-facing polish; it’s about helping reps communicate more clearly so things don’t get lost in translation. It saves time, reduces friction, and keeps internal operations running smoothly.

Final proposals and pricing conversations

Once you’re in the proposal stage, every word carries weight. These documents are often the last thing a buyer sees before making a decision. They need to be specific, accurate, and completely aligned with what’s been said during the sales cycle.

AI will always be generic. That’s what it’s built to do. You can’t ask it to capture the nuance of a buyer’s situation or the exact tone of a months-long conversation. Even a small disconnect can raise doubts or create friction right at the finish line.

If your team is using AI to write or edit proposal language, stop them. Those words need to come from someone who knows the customer, the offering, and the relationship you’ve built. If AI is used at all, it should be in a draft-only capacity and followed by a human rewrite, line by line.

RFP answers that rely on company experience or case-specific history

There’s a difference between responding to “What industries do you serve?” and “Tell us about your results with a company like ours.” One is fair game for templating. The other demands a real, contextual answer.

When your team receives an RFP or a custom questionnaire, they need to treat those responses like they’re being read by someone who will notice if the answer feels robotic. And chances are, they will.

It’s fine to use AI to help reword boilerplate or speed up responses to general questions.

But if the question touches on your track record, your unique capabilities, or your experience in a niche setting, AI should stay out of it entirely. Those answers need to be written by the people who actually know the story.

Mid-to-late stage deal communication

As a deal progresses, trust becomes the most valuable currency. The further down the funnel you go, the more precise your communication needs to be. You’re responding to objections, navigating internal politics, and getting into the weeds with decision-makers.

AI isn’t equipped for that. It doesn’t know your buyer as a person. It doesn’t know the personalities, the priorities, or the pressure that’s building around the deal. And your prospect can tell when your responses feel scripted or off-target.

If your reps are leaning on AI to answer buyer emails late in the process, it’s time to have a serious conversation. This is where human connection and judgment matter most. If they aren’t confident in what they’re writing, they shouldn’t reach for AI; they should be reaching out to you.