When it comes to sales, the winners use specific tactics that give them the best chance at winning.

Nowhere is that more apparen than in one of the most important parts of the process – the sales call. It often determines if you win or lose the deal.

Whether it’s a discovery call, a demo, or a deep technical review, winning sales reps must be able to show their expertise in understanding the customers’ problem.

How do the best reps do this?

By leveraging AI to prepare for sales calls, sound knowledgable, and perform better in-person. Here are 5 ways you can use AI to win sales calls in 2025:

1. Understand What’s Happening with Your Prospect Right Now

You’d be surprised how many reps don’t do this. Researching your prospect is low-hanging fruit. It’s not hard to figure out what’s happening with your client’s company.

And when you know what the company is going through, you can speak to those pain points. The worst thing you could do is jump on a call and be clueless about some big event, like going public or merging with a competitor. 

Tools to Use: Perplexity, Claude, Microsoft Copilot, Gemini, or Anthropic. Ask questions about the company, news, recent blogs, posts, etc. This will give you a sense of any big changes, like a big hiring spree or, in contrast, recent layoffs. 

Triangulate: Don’t limit yourself to one tool for research. Don’t rely solely on Google News, or just ask ChatGPT what’s going on, because you run the risk of getting outdated information. Instead, ask a mix of AI tools like Perplexity, Claude, and Gemini. See what kind of responses you get from 2-3 sources, and then compare your notes. 

Social Media: You’ll also want to use tools like Sprout Social with AI Assist to analyze your prospect’s socials. This does have limitations for videos, so for YouTube and TikTok, it may have to be more manual. See what they are announcing on Instagram, what kinds of articles are they releasing on LinkedIn, and what they’re putting out on X. This can give you a quick look into the hottest topics from their perspective, so you can speak their language when you pick up the phone. 

Why This Matters: Timing is everything in sales. Understanding the current state of your customer can give you a strong edge (or help you avoid a disastrous pitch).

You don’t want to push a product if the company is suffering and you don’t have the answer to their problems. Likewise, you don’t want to waste time offering upsells or services they don’t even need. 

When you call with context, you show your prospect that you understand what’s happening inside the company right now and that you have answers to their problems. This information is especially helpful when you’re talking to a lead for the very first time.

2. Gain Insights into Previous Interactions

Unless it’s a brand new prospect, you likely have some background information already in your company files. Dig into your CRM and find those past interactions. Again, you’ll be surprised how many reps don’t do this basic step.

Crucial details could be trapped in emails.

Internal chat logs. 

Call transcripts. 

Compile any notes you can find on those exchanges. There’s a wealth of information here for you. 

But it can be hard to find fragmented knowledge. This is where AI comes in. AI tools like Gemini in Gmail or Gong for Zoom call transcripts can pull insights for you and surface them rapidly, so you’ve got everything you need right in front of you. 

This is especially helpful when: 

• You just inherited a new account. 
• You’re assigned a deal that’s been sitting in the system. 
• The prospect has been in your CRM for a while, but it’s never closed. 

You’ve got a few key questions to ask AI to answer from those past interactions: 

• What’s the current engagement level of this prospect? 
• Have we been spamming the client with no response? 
• Are they asking for demos and seeking more information? 

AI will help you condense transcripts and heavy note-taking to get to the important points. 

Here are examples of insights you can find buried in the CRM:

• Champions who liked your product, but didn’t have the budget for it <-Bad timing?
• Past demos with questions or concerns that you’ll have to address on the next call. <- Objection handling
• Competing vendors, what they offer, and how your product compares. <- Be ready for a bakeoff
• Any pushback on security or compliance issues. <- Might want to loop in your security team

And perhaps most important of all: you can surface past blockers to the deal. If a client pushed back against something in your product that they didn’t understand or admire, you’ll need to come prepared with an answer. 

Why It Matters: Having this information from past interactions allows you to come prepared, with guns blazing, so you don’t repeat the mistakes made in the past. The ones that have stopped the client from pulling the trigger until now. 

3. Prepare Answers Your Prospect WANTS to Hear

Now that you’ve got your news and your CRM history, you can use AI to help you with the actual sales calls. 

To be clear: you should NOT copy anything AI gives you word-for-word. You are human, after all, and your prospect does not want to talk to a robot.

Instead, you’ll leverage AI to generate short talk tracks with one-liners or bullet points, so you’re ready with important talking points. 

For example, you could use a tool like 1up to generate bullets from your news and your CRM history with a prompt like this: 

The result should be a quick, contextual talk track you can use to make a call right now. 

Tools to use: 1up, Microsoft Autopilot, Anthropic, or Gemini. 

Why It Matters: AI is great for specific, contextual inputs. You don’t want a generic talk track because it will just be robotic, but AI can help you with “less is more.” You’ll get short, sharp, buyer-specific notes to prepare for your call. They’ll serve as reminders while you’re on the phone. 

4. Be Ready for Objections in Real Time

Being ready for objections is EVERYTHING. 

You might be on the phone with a prospect, and they give you an objection. 

For example: 

“Hey, I like what you guys do, but I don’t think we’re ready for a cloud-hosted ready solution. Our company is pretty cautious when it comes to sharing information in the cloud.” 

In the past, you’d have to get off the call. You’d have to sift through files and find someone who can help you formulate a reasonable, clear response that will overcome those fears. It could take you hours or maybe even days to get back to the prospect. And by then, your competitor may have an edge over you.

But today, you can simply ask an automated knowledge base for a response. 

You can respond, in seconds, to your prospect, “Good point. Let me give you some feedback on your concerns. Give me just a second…” 

You can then punch in the objection to a tool like 1up, and you’ll get an instant response directly from your knowledge base to provide to your prospect. 

This ensure you’re relaying over the most accurate information in the fastest time possible.

Why It Matters: You now look professional and credible. You look ready for anything. And you look like the smartest person in the room. And you never had to get off the phone or leave your keyboard. 

Here’s what AI-powered objection handling looks like in practice:

5. Follow Up… But Make it Personal

Anyone can send an email. And almost everyone is using AI to draft a “personalized” email. 

But, let’s be honest: those emails suck. 

They all sound the same. And the customer knows it. 

So: don’t do that. Be different.

Do not copy/paste an AI-generated email. It’s one of the many ways automation is hurting sales. 

If your goal is to stand out, you need to do the opposite. 

Instead, use AI to generate reference posts, to get critical information, to get news, and to generate a transcript from your call. AI can help you refresh your memory from your call and give you bullet points.

But write the email yourself with your personality and specific references to your conversation.

Why It Matters: You have established a strong relationship with the client up to this point. You want to build on that relationship, not destroy it. Remember, AI is your assistant, not your replacement. 

Bonus Ideas: How NOT to Use AI When Prepping for Calls

1. Don’t generate a sales script and read off it. Reading off scripts is bad. Reading off AI slop is 10x worse.
2. Don’t rely solely on AI to give you company news. Remember, some model knowledge is outdated by several months. What you want is to get multiple sources of truth (Google, Bing, Copilot, Gemini, etc)
3. Don’t use AI to tell you a sales strategy. Yes, there are tools that promote this. But it’s not a reasonable expectation for sales reps.

At the end of the day, you’ll need to know how to close the deal on your own. Asking an AI for this information is going to give you notes you already have.