Sales is a job where the gap between your highest-value activities and everything else is massive. The things that actually close deals, like building relationships, understanding a prospect's real pain points, and running a sharp demo, require all of your focus. But the reality of the day-to-day is that you spend a huge chunk of time on research, email follow-ups, CRM updates, pipeline reviews, proposal drafting, and pulling together reports. Every hour spent on that kind of work is an hour you're not spending with customers.

That's where AI agents come in. This guide walks through the major responsibilities of a sales professional and shows you how AI agents can take over the busywork so you can spend your time where it actually moves the needle.

An important note on ethical AI use

At AInalysis, our mission is to empower individuals with the tools and knowledge they need for the artificial intelligence future. That means helping you become more effective, not replaceable. The goal of using AI in sales is to handle the time-consuming research, writing, and analysis so you can focus on the work that actually requires your relationship skills, product knowledge, and deal instincts. Those are things AI simply cannot replicate.

Every suggestion in this guide is designed to keep you in control. You review everything, you make the final call, and you bring the human connection that wins trust and closes deals.

Prerequisite: What are AI agents and how are they different from chatbots?

Before we get into the specifics, it's worth understanding what we mean by "AI agent" throughout this page. If you've used a chatbot that follows a script and gives canned responses, that's not what we're talking about here.

AI agents like Claude Cowork are general-purpose tools that can understand context, reason through multi-step problems, browse the web, create documents, draft original content, and adapt their output based on what you ask for. They don't follow a script. You give them a task in plain language and they work through it.

For a deeper breakdown of the difference between traditional chatbots and the new generation of AI agents, check out our guide on What are AI agents?.

More on AI agents before we start

If you instead want to see a high-level overview of what AI agents can typically do (before reading this page), check out our guide on How to use AI agents: 7 powerful use cases with example prompts.

Also, if you do not have access to an AI agent yet, this guide will walk you through how to get started with one and start putting it to work.

The key takeaway: AI agents are flexible enough to help with many tasks you do in your role. The examples below will show you exactly how.

Prospecting and lead research

Before you ever reach out to a prospect, you need to know who they are, what their company does, what challenges they're likely facing, and why your solution might matter to them. Good prospecting isn't just finding names and emails. It's building a picture of the account that lets you show up to the conversation informed and relevant. The problem is that kind of research takes serious time, especially when you're doing it for dozens of prospects every week.

AI agents can do the research for you. They can browse a prospect's company website, read their recent news and press releases, check their LinkedIn activity, and pull it all together into a structured profile you can scan before you pick up the phone or hit send on an email.

Example prompts:

"I have a list of 20 target accounts in a spreadsheet called 'target-accounts-q1.csv' on my desktop. It has the company name, website URL, and the contact name for each. Browse each company's website and find their company size, industry, what products or services they sell, and any recent news or announcements. Add all of this information to each row and save the enriched file as 'target-accounts-q1-enriched.csv' on my desktop."

"I have a discovery call tomorrow with the VP of Operations at Acme Manufacturing. Browse their website at acmemanufacturing.com and search the web for any recent news, press releases, or executive interviews. Then read our product one-pager 'product-overview.pdf' on my desktop. Create a pre-call research brief on my desktop that includes: company overview, key people I should know about, likely pain points based on their industry and size, how our product aligns with those pain points, and three smart questions I can ask during the call."

"Search the web for mid-sized logistics companies in the northeast United States that have been in the news recently for supply chain challenges or technology upgrades. I'm looking for companies that might benefit from our fleet management software. Compile a list of 10 strong prospects with their company name, headquarters, estimated company size, a summary of the relevant news I should reference in my outreach, and the best person to contact. Save it as a prospecting list on my desktop."

Outreach and follow-up emails

Personalized outreach is what separates the emails that get replies from the ones that get deleted. But writing a genuinely personalized email for every prospect on your list is one of the most time-consuming parts of sales. You know the generic templates don't work, yet you don't have three hours a day to hand-craft every message.

AI agents can bridge that gap. With the Gmail connector, an agent can research a prospect, draft a personalized email referencing something specific about their company, and have it sitting in your Gmail drafts ready to send. It's not a template with a name swapped in. The agent actually reads the prospect's website and tailors the message.

A note on connectors: The first time you ask an agent to access your Gmail, it will prompt you to set up the Gmail connector. This is a one-time authorization step where you grant the agent permission to read and draft emails on your behalf. Once you approve it, the connector stays active and you can use it in any future prompt without setting it up again.

Example prompts:

"Read the spreadsheet 'demo-attendees-feb.csv' on my desktop. These are 12 people who attended our product demo webinar yesterday. The file has their name, email, company, and title. Browse each person's company website to understand what they do. Then draft a personalized follow-up email in Gmail for each attendee that thanks them for joining, references something specific about their company that connects to what we showed in the demo, and includes a CTA to book a 15-minute call. Leave all 12 as drafts so I can review and send."

"Search my Gmail for the last email thread with sarah.johnson@techstartup.io. Read through the full conversation to understand where we left off. Then browse their company website to check for any recent news or product updates I can reference. Draft a follow-up email in Gmail that naturally picks up the conversation, mentions something current about their business, and moves toward scheduling a next call. Leave it as a draft."

"I have a list of 8 deals that went cold over the past 60 days saved as 'stalled-deals-q1.csv' on my desktop. It has the contact name, email, company, and the last note from our conversation. For each contact, browse their company website to find something new I can reference, then draft a re-engagement email in Gmail that feels natural and timely, not like a generic 'just checking in' message. Reference the specific thing we were discussing and tie it to something current about their business. Leave all 8 as drafts."

Meeting preparation and call planning

Walking into a sales call unprepared is one of the fastest ways to lose a deal. The best reps do their homework: they know the company, they know the person they're talking to, they've reviewed the deal history, and they have a game plan for the conversation. But that prep work can easily eat 20-30 minutes per call, and when you have four or five calls a day, that adds up.

AI agents can do the prep for you in minutes. They can browse the prospect's website, search for recent news, read through your existing notes and deal files, and compile everything into a structured call prep document with talking points and questions ready to go.

Example prompts:

"I have three discovery calls this afternoon. The companies are listed in the file 'todays-calls.txt' on my desktop with their website URLs and the contact person's name and title. For each call, browse the company's website and search the web for any recent news about them. Then create a single call prep document with a section for each company that includes: company overview, the contact's likely priorities based on their title, three to five discovery questions tailored to their business, and potential objections I should be ready for. Save it on my desktop."

"I'm about to join a call with a prospect who's been evaluating us for six weeks. Read the deal file 'acme-corp-deal-notes.txt' on my desktop, which has all my notes from previous conversations. Then browse their website to check if anything has changed since we last spoke. Search the web for any recent news about their company or industry. Create a call prep sheet that summarizes the deal history, highlights the prospect's key concerns from past conversations, and gives me three strategies for moving toward a close. Save it on my desktop."

"I have a big presentation to the C-suite at a healthcare company tomorrow. Read the file 'healthco-proposal-draft.docx' on my desktop. Then browse their website at healthco.com, read their most recent annual report if it's publicly available, and search for any recent interviews with their CEO or CTO. Create a detailed prep document that includes: executive bios, their stated strategic priorities, how our solution maps to those priorities, potential tough questions they might ask, and suggested answers. Save it on my desktop."

Proposal and presentation creation

Proposals and presentations are where deals get won or lost. A strong proposal directly addresses the prospect's specific pain points, maps your solution to their needs, and makes the business case clearly. But building these from scratch for every deal is a grind, especially when you need to pull in competitive positioning, pricing details, and relevant case studies.

AI agents can assemble the pieces. They can read your deal notes, browse the prospect's website for context, pull in relevant information from your existing collateral, and produce a draft proposal or presentation outline that's already tailored to the opportunity.

Example prompts:

"Read my deal notes in 'meridian-deal-notes.txt' on my desktop and the prospect's requirements document 'meridian-rfp-response-needed.pdf'. Then browse meridian-group.com to understand their business and read our case studies file 'case-studies-2025.pdf' on my desktop. Draft a proposal document that includes: an executive summary tailored to Meridian's stated needs, a solution overview mapping our features to their requirements, the two most relevant case studies from companies similar to theirs, and a recommended pricing structure based on what they described. Save it as 'meridian-proposal-draft.docx' on my desktop."

"I need a presentation deck outline for a mid-stage deal. Read the files 'summit-logistics-deal-notes.txt' and 'product-overview.pdf' on my desktop. Then browse summit-logistics.com and search the web for recent news about challenges in the logistics industry. Create a 10-slide presentation outline where each slide has a headline and four to five bullet points. The narrative should go: their challenge, the cost of doing nothing, our solution, how it works, relevant results from similar customers, implementation timeline, and next steps. Save the outline on my desktop."

"Read the file 'competitor-comparison.xlsx' on my desktop, which has a feature-by-feature comparison of our product vs. three competitors. Then browse each competitor's website to check if any of the information is outdated (new features, pricing changes, etc.). Update the spreadsheet with any new information you find and add a 'Last Verified' date column. Save the updated file on my desktop so I have an accurate comparison ready for my next deal review."

Pipeline management and forecasting

A healthy pipeline is everything in sales. You need to know what's in it, what's real, what's at risk, and what the numbers actually say about whether you're going to hit your target. But pipeline reviews often turn into an exercise of scrolling through rows of CRM data trying to spot patterns, identify stalled deals, and figure out where your gaps are. That analysis is crucial, but it doesn't need to take an hour every time.

AI agents can read your pipeline data, run the analysis, and surface the insights you need to make decisions. They can flag deals that haven't moved, identify gaps in your coverage, and even help you build a more accurate forecast.

Example prompts:

"Read the spreadsheet 'pipeline-export-feb.csv' on my desktop. This is a full export of my current pipeline with deal name, stage, value, days in stage, last activity date, close date, and probability. Analyze the data and flag every deal that's been in the same stage for more than 21 days with no activity. For those stalled deals, categorize the likely risk level (low, medium, high) based on how long they've been stuck and their deal size. Then create a pipeline health report that includes: total pipeline value, weighted pipeline, stage-by-stage breakdown, a list of at-risk deals with recommended next actions, and a gap analysis showing how much more pipeline I need to build to hit my $2M quarterly target at our historical 30% close rate. Save the report on my desktop."

"I need to prepare my weekly forecast for my manager. Read the file 'pipeline-export-feb.csv' on my desktop. For each deal in the 'Negotiation' or 'Proposal Sent' stage, assess the close probability based on deal size, days in stage, and last activity date. Create a forecast document with three scenarios: conservative (only deals with 80%+ probability), likely (deals with 50%+ probability), and optimistic (all deals weighted by their probability). Include the total dollar amount for each scenario and a brief note on the two or three deals most likely to swing the number. Save it on my desktop."

"Read two spreadsheets on my desktop: 'pipeline-jan.csv' and 'pipeline-feb.csv'. Compare the two to identify which deals advanced stages, which went backward, which were lost, and which are new. Calculate the month-over-month movement in total pipeline value, average deal size, and win rate. Write a pipeline trend analysis that highlights the key changes and any patterns I should be paying attention to. Save it on my desktop."

Competitive intelligence and battle cards

Knowing what your competitors are doing, what they're charging, and how they're positioning themselves is the difference between handling objections smoothly and getting caught flat-footed. But keeping your competitive intelligence current requires constant monitoring, and most reps just don't have time to check competitor websites and read industry news every week.

AI agents can do the monitoring for you. They can browse competitor websites, search for recent news and product updates, and produce updated battle cards and competitive briefs that you can use in your next call.

Example prompts:

"Browse the websites of our three main competitors: competitorA.com, competitorB.com, and competitorC.com. For each, check their pricing page, features page, and any recent blog posts or press releases. Then search the web for any recent news, analyst reports, or customer reviews mentioning these companies. Create an updated competitive battle card for each competitor that includes: their current positioning, pricing, key differentiators, recent moves, known weaknesses, and how we should position against them. Save it as 'competitive-battle-cards-feb.docx' on my desktop."

"Read the file 'lost-deal-notes-q4.csv' on my desktop. This has the deal name, competitor we lost to, and the reason we lost for every deal we didn't win last quarter. Analyze the data to identify which competitors we're losing to most often and the top reasons why. Then browse those competitors' websites to see if they've made any recent changes that address those winning themes. Write a competitive loss analysis report with actionable recommendations for how we can improve our win rate against each competitor. Save it on my desktop."

"I just heard that CompetitorB launched a new pricing tier. Browse competitorB.com/pricing and read through the details. Then read our current pricing document 'our-pricing-2026.pdf' on my desktop. Create a side-by-side comparison that highlights where we're more competitive, where they undercut us, and the key talking points I should use when a prospect brings up their new pricing. Draft an email in Gmail to our sales team summarizing the change and sharing the talking points. Leave it as a draft for me to review."

Sales reporting and performance analysis

Whether you're preparing for a quarterly business review, reporting to your manager, or just trying to understand your own numbers, sales reporting requires pulling data together from multiple sources and turning it into a clear story. Most of the time, the data is already there in spreadsheets and CRM exports. The challenge is doing the analysis and building the narrative around it.

AI agents can read your raw data, do the calculations, spot the trends, and produce a formatted report or presentation outline that's ready to share.

Example prompts:

"Read the spreadsheet 'q1-sales-results.xlsx' on my desktop. This has every closed deal from last quarter with the rep name, deal value, close date, sales cycle length, product sold, and lead source. Analyze the data by rep, by product, and by lead source. Calculate total revenue, average deal size, win rate, and average sales cycle length for each breakdown. Then create a quarterly sales performance report with the key metrics, top performers, areas for improvement, and a comparison to our Q1 targets. Save it on my desktop."

"I need to prepare for my QBR with leadership. Read the files 'q1-sales-results.xlsx', 'q1-pipeline-summary.csv', and 'q1-activity-metrics.csv' on my desktop. Create a presentation outline for a 30-minute review that covers: revenue achieved vs. target, pipeline generation and health, activity metrics, top wins, key losses and lessons learned, and my plan for next quarter. For each slide, write the headline and the key talking points with specific numbers from the data. Save the outline on my desktop."

"Read the spreadsheet 'rep-activity-log-feb.csv' on my desktop. This tracks daily activities for each rep on my team: calls made, emails sent, meetings held, and proposals delivered. Calculate the average activity levels per rep and identify who's above and below the team average in each category. Then cross-reference with 'pipeline-export-feb.csv' to see if there's a correlation between activity levels and pipeline generation. Write up the findings in a coaching summary that I can use in my one-on-ones this week. Save it on my desktop."

Account management and expansion

For most sales organizations, expanding revenue within existing accounts is just as important as winning new ones. But account expansion requires staying on top of what's happening with your customers: understanding their evolving needs, spotting upsell opportunities, and making sure renewals don't slip through the cracks. When you're juggling dozens of accounts alongside new business prospecting, it's easy for expansion opportunities to get lost.

AI agents can help you stay on top of your accounts by reading through your customer data, identifying patterns that signal an upsell or renewal risk, and preparing the outreach you need to act on those signals.

Example prompts:

"Read the spreadsheet 'customer-accounts.xlsx' on my desktop. This has every active customer with their contract value, renewal date, product tier, and usage notes from our last QBR. Identify all accounts with renewals coming up in the next 90 days and flag any that have had support escalations or declining usage (noted in the usage column). For each upcoming renewal, draft a personalized check-in email in Gmail that references their specific situation and proactively addresses any concerns. Leave them all as drafts so I can review and send."

"Read the file 'customer-usage-data-jan.csv' on my desktop. This has usage metrics for each customer account: logins per month, features used, and support tickets filed. Identify accounts that are only using a fraction of the features available on their plan, because those are candidates for either a training session or an upgrade conversation depending on the situation. Create a document with two lists: accounts that might benefit from a success check-in (underusing their current plan) and accounts that might be ready for an upgrade (maxing out their current plan). Save it on my desktop."

"I have a big renewal coming up with one of our enterprise accounts. Read my account notes in 'globaltech-account-history.txt' on my desktop and the file 'globaltech-usage-report.csv'. Then browse globaltech.com to check for any recent company news, leadership changes, or strategic shifts. Create an account review document that summarizes their history with us, highlights how their usage has grown, identifies which additional products they'd benefit from, and includes talking points for positioning an expanded deal. Draft an email in Gmail to my contact there requesting a renewal review meeting. Leave it as a draft."

Getting started

You don't need to overhaul your entire workflow to start benefiting from AI. Start with one or two areas where you spend the most time on repetitive work. For most sales professionals, that's prospecting research and email follow-ups.

Get started with an AI agent like Claude Cowork, hand it your prospect list or a deal file, and see what it produces. You'll quickly get a feel for how to phrase your prompts and where the tool is most helpful for your specific workflow.

From there, you can expand into pipeline analysis, competitive intelligence, reporting, and everything else covered in this guide. The key is to start small and build the habit.

Want to learn more about AI agents and what they can do? Check out our guides on AI agent use cases or explore our complete library of AI resources.