Walk In Prepared: Own Your Discovery Call

The best sales reps walk into every call knowing more than the prospect expects. They know the company's recent moves, the buyer's likely priorities, the competitive landscape, and the questions that will open up real conversation. AI doesn't replace that instinct, it compresses the research time from two hours to fifteen minutes and surfaces angles you would have missed. This lesson is your reference guide for using AI tools to prepare for sales calls and run sharper discovery.

7 Things to Know Before You Start

1. AI tools like ChatGPT Plus and Claude Pro can synthesize public information about a company or industry in seconds, but they don't have access to your CRM or private data unless you paste it in.
2. Your prompts are your instructions. The more context you give the AI, the better the output. A vague prompt gets a generic answer; a specific prompt gets a usable one.
3. AI-generated research can contain outdated or incorrect details, especially on company financials, leadership, and recent news. Always verify facts before the call.
4. Microsoft Copilot (inside Teams and Outlook) can pull from your company's internal files and emails if your organization has it configured, other tools cannot.
5. Discovery questions are one of the highest-value outputs AI can produce for sales prep. You can generate 20 tailored questions in under two minutes.
6. AI can role-play as a skeptical buyer. Use it to rehearse objection handling before you pick up the phone.
7. Everything you type into a public AI tool (ChatGPT, Claude, Gemini) is potentially used for training unless you opt out or use an enterprise version. Never paste in confidential client data.

Concept 1: What AI Can Actually Research for You

Before a call, most reps skim a company's LinkedIn page and maybe glance at their website. That takes 20–30 minutes and still leaves gaps. AI tools, particularly ChatGPT Plus with Browse enabled, or Perplexity AI, can pull together a structured company brief in under five minutes. You get a summary of what the company does, their market position, recent news, known pain points in their industry, and even the typical buying concerns of someone in your contact's role. That's a meaningful head start before you've even opened your CRM.

The key is treating AI like a research assistant who needs a clear brief, not a search engine you throw keywords at. When you tell it your contact's title, their company size, their industry, and the product you're selling, it can build a prep document that feels like something a sales engineer spent an afternoon on. You still bring the relationship and the judgment. AI brings the raw material. Used well, this shifts your prep time from information-gathering to actual thinking about strategy.

• Company overview: what they do, who they serve, approximate revenue range, headcount
• Recent news: funding rounds, product launches, leadership changes, acquisitions, layoffs
• Industry trends: what's pressuring or exciting their sector right now
• Buyer role context: typical priorities, KPIs, and frustrations of your contact's job title
• Competitor landscape: who else they might be evaluating in your category
• Trigger events: things that might make them more likely to buy right now
• Likely objections: common pushback from companies at this stage or size

Click to enlarge

Concept 2: Building Tailored Discovery Questions with AI

Discovery is where deals are won or lost. Generic questions, "What does your process look like today?", get polite but shallow answers. Specific, insight-led questions signal that you've done your homework and unlock real conversation. The problem is that writing genuinely tailored discovery questions for every call takes time most reps don't have. AI solves this. Feed it the buyer's role, company context, your product category, and the deal stage, and it will generate 15–20 sharp questions in about 60 seconds.

The output quality depends on how much context you provide. A prompt that says "give me discovery questions for a sales call" produces textbook filler. A prompt that says "give me 10 discovery questions for a call with a Director of Marketing at a 300-person B2B software company, who currently uses HubSpot, and I'm selling a LinkedIn advertising management platform" produces questions you could actually use on Monday. The AI is pattern-matching against thousands of similar sales scenarios, your job is to give it enough signal to match the right patterns.

1. Include the buyer's exact job title, "Marketing Manager" and "CMO" get very different questions
2. Specify company size (headcount or revenue range), priorities shift dramatically at 50 vs. 500 employees
3. Name the industry, a question that works for a healthcare buyer falls flat with a retail buyer
4. Mention any known tools or vendors they currently use, questions about switching cost and integration land better
5. State your product category clearly, don't just say "our software," say "project management software" or "sales forecasting tool"
6. Tell the AI the deal stage, first call questions differ from second-call or late-stage questions
7. Ask for a mix of question types: situational, problem-focused, implication, and vision questions
8. Request that the AI flag which questions are highest priority if time is short

Concept 3: Using AI to Anticipate Objections Before the Call

Experienced reps know that objections are predictable. A CFO will push on price and ROI. An IT stakeholder will ask about security and integration. A mid-level manager will worry about change management and their team's bandwidth. The specific objections vary by industry, deal size, and buyer profile, but they're rarely a surprise if you've thought them through in advance. AI is exceptionally good at generating likely objections for a given buyer profile, and then helping you prepare crisp, confident responses before the call.

The most useful format is to ask the AI to both generate the objection and draft a response you can adapt. This turns your pre-call prep into a mini rehearsal. You can also use AI to role-play the buyer, you respond to the objection in the chat, and the AI plays the skeptical prospect pushing back. This is particularly effective for handling objections you've historically struggled with, or for high-stakes calls where you can't afford to be caught flat-footed. Think of it as a flight simulator for the conversation.

Part 1 Cheat Sheet

• AI can cut call prep time from 60–90 minutes to 10–15 minutes, if you give it specific context
• Best tools for research: ChatGPT Plus with Browse enabled, Perplexity AI, Google Gemini
• Best tools for question generation and objection prep: Claude Pro, ChatGPT Plus
• Always include: contact's exact title, company size, industry, your product category, and deal stage
• Ask for a mix of question types: situational, problem-focused, implication, and vision
• Use AI to generate likely objections by buyer type, then draft your responses in advance
• Never paste confidential client data into public AI tools, use enterprise versions for sensitive info
• AI output is a starting point, verify facts, verify numbers, and put responses in your own voice
• Save a master prep prompt template, reuse it for every call by swapping in new details
• Microsoft Copilot inside Teams/Outlook can access your internal files, all other tools cannot

Key Takeaways from Part 1

• AI compresses research time dramatically, the value is in what you do with the time you get back
• Specificity is everything. Generic prompts produce generic output. Detailed prompts produce usable briefs.
• Discovery questions are one of the highest-ROI things AI can generate for a sales professional
• Objection prep before the call is more effective than improvising on the call. AI makes this fast
• Treat every AI output as a draft that you refine, not a finished product you deliver

Part 1 covered the foundation, why AI belongs in your pre-call routine and how to pull basic prospect intelligence. Now the work gets more specific. This section is your operational reference: which AI tools handle which tasks, how to structure discovery questions using AI, and how to avoid the most common mistakes sales professionals make when they trust AI output without verifying it. Keep this section open during your prep sessions.

7 Things Every Sales Pro Should Know About AI-Assisted Call Prep

1. AI does not browse the internet by default. ChatGPT's free version has a knowledge cutoff, meaning it won't know about last quarter's earnings or a company's recent layoffs unless you paste that information in yourself.
2. Prompt quality determines output quality. Vague prompts produce generic responses. Specific prompts, with company name, role, deal context, and your goal, produce usable prep materials.
3. AI can generate discovery questions, but you must filter them. A 20-question list from AI is a starting point, not a script. Pick 5-7 that fit your specific deal.
4. Real-time tools like Perplexity AI, ChatGPT with web browsing (Plus tier), and Microsoft Copilot can pull current news, recent press releases, and LinkedIn activity, use these for live prospect research.
5. You can paste documents directly into Claude Pro or ChatGPT Plus, annual reports, RFPs, LinkedIn profiles copied as text, and ask the AI to summarize, flag risks, or generate questions based on that content.
6. AI-generated personas are useful but approximate. A 'VP of Sales at a 200-person SaaS company' prompt gives you a useful archetype, not a portrait of your actual buyer. Combine it with what you already know.
7. Confidentiality matters. Never paste client contracts, internal pricing, or sensitive CRM data into a public AI tool. Use enterprise versions with data privacy agreements when handling sensitive deals.

Building a Buyer Profile Before the Call

A buyer profile is a one-page summary of who you're talking to, what they care about, and what pressures they're under. Traditionally, building one meant 30-45 minutes of scattered research across LinkedIn, Google News, and the company website. With AI, you can compress that to 8-12 minutes, not by skipping research, but by processing it faster. You gather the raw inputs (LinkedIn bio, company page, recent news), paste them into your AI tool, and ask for a structured summary with talking points.

The most effective buyer profiles for sales calls include four layers: the person (role, tenure, background, likely priorities), the company (size, industry, recent moves, known challenges), the deal context (why they might be looking, what they've tried before, budget signals), and the competitive landscape (who else they're likely evaluating). AI can draft all four layers in under two minutes if you feed it the right inputs. The output won't be perfect, but it gives you a structured starting point you can refine in the time it used to take to find the company's LinkedIn page.

• Paste the prospect's LinkedIn summary + company 'About' page into ChatGPT or Claude and ask: 'Summarize this person's likely priorities and pressures as a [role title].'
• Ask AI to identify 3 business challenges common to companies of this size and industry, then verify which apply before the call.
• Use Perplexity AI or ChatGPT with browsing to search '[Company name] news 2024' and ask for a 5-bullet summary of recent developments.
• Ask AI: 'What questions would a [job title] at a [industry] company most want answered before buying [your product category]?'
• Request a 'potential objection list' based on the company profile, then prepare one crisp response to each before you dial.

Click to enlarge

AI Tools for Call Prep: What Each One Does Best

Click to enlarge

Generating Discovery Questions That Actually Work

Discovery questions are the backbone of a consultative sales call. The problem is that most salespeople walk in with the same 10 questions they've used for three years, or they freestyle and miss critical areas. AI can generate a tailored question bank in under 60 seconds. The key is giving it enough context: your product, your prospect's role and industry, the stage of the deal, and what you already know about their situation. Generic input returns generic questions. Specific input returns questions that make prospects say 'That's exactly what we're dealing with.'

Structure your AI-generated question bank across four discovery categories: situation (what's their current state), problem (what's broken or inefficient), implication (what does that cost them in time, money, or risk), and need-payoff (what would solving it mean for them). This mirrors the SPIN Selling framework and maps directly to how buyers make decisions. Ask AI to generate 4-5 questions per category, then select the most relevant ones for your specific call. You'll arrive with a sharper question set than you'd build manually, and in a fraction of the time.

1. Open ChatGPT or Claude and start with context: 'I'm a [your role] at a [your company type] selling [product/service]. I'm about to call a [prospect role] at a [company type].'
2. Add what you know: 'They recently [trigger event, e.g., hired a new CRO / expanded to Europe / had a data breach].'
3. Request by category: 'Generate 4 situation questions, 4 problem questions, 3 implication questions, and 3 need-payoff questions for this discovery call.'
4. Ask for a follow-up layer: 'For each problem question, give me one follow-up probe if they answer yes.'
5. Filter ruthlessly, pick the 6-8 questions that feel most relevant to this specific deal and delete the rest.
6. Ask AI: 'Which of these questions is most likely to uncover budget authority or decision-making process?', use that to prioritize your sequence.
7. Save the full question bank as a Notion page or Google Doc template organized by deal type for reuse.

Click to enlarge

Prompt Example: Full Pre-Call Brief

Handling AI Hallucinations and Accuracy Risks in Sales Prep

AI tools sometimes state incorrect information with complete confidence. This is called hallucination, the model generates plausible-sounding content that isn't accurate. In sales prep, this is a real risk. If AI tells you a prospect's company raised $50M in Series B funding and that number is wrong, you may reference it on the call and damage your credibility instantly. The risk is highest when you ask AI to research facts it can't verify, company revenue, headcount, funding rounds, executive tenure, recent news, without giving it the source material directly.

The fix is a two-step habit: generate with AI, verify with primary sources. Use AI to synthesize and structure the research you've already gathered, not to invent the facts from scratch. For anything you plan to reference on a call, a statistic, a company detail, a trigger event, confirm it on the company's website, LinkedIn, a press release, or a credible news source before you say it out loud. AI is your research assistant, not your fact-checker. That distinction will save you from an embarrassing moment mid-call.

Click to enlarge

Practice Task: Build a Complete Pre-Call Brief

Quick Reference: AI Call Prep Cheat Sheet

• Paste, don't ask from scratch, give AI real source material to analyze rather than asking it to invent facts about a company.
• Use Perplexity AI or ChatGPT with browsing for real-time research; use Claude Pro for analyzing long documents you supply.
• Structure discovery questions in four layers: Situation → Problem → Implication → Need-Payoff.
• Always ask AI for a 'likely objections' list, then prep one crisp response to each before you dial.
• Verify any specific number, funding detail, or recent event before referencing it on a call.
• Never paste sensitive client data, internal pricing, or contract terms into a public AI tool.
• Use Microsoft Copilot inside Outlook or Teams to summarize existing email threads before a follow-up call, it's faster than re-reading everything.
• Build reusable prompt templates for your most common call types (new logo discovery, renewal QBR, competitive displacement) and save them in Notion or Google Docs.
• AI question banks are a starting point, always filter down to 6-8 questions that fit this specific deal and prospect.
• Post-call: paste your notes into AI and ask it to identify gaps in your discovery and suggest what to ask next time.

Key Takeaways from Part 2

• Different AI tools serve different prep tasks, match the tool to the job rather than defaulting to one platform for everything.
• The SPIN framework gives your AI-generated questions a structure that mirrors how buyers actually make decisions.
• AI hallucination is a real risk in sales prep, use a generate-then-verify habit for any factual claims you plan to reference.
• The most powerful use of AI in call prep is synthesis: you gather the raw inputs, AI organizes them into a usable brief in minutes.
• Reusable prompt templates and saved question banks turn one-time AI work into a permanent competitive advantage across your entire pipeline.

The sharpest sales reps don't wing discovery calls, they engineer them. AI gives you a repeatable system for walking into every call with the right questions, the right context, and a clear hypothesis about what the prospect actually needs. This section is your reference guide for turning that system into a daily habit.

1. AI cannot access a prospect's internal data, it works from public information only.
2. Always verify AI-generated company facts before the call. Funding figures and headcount change fast.
3. The best discovery prompts include the prospect's role, industry, company size, and your product category.
4. AI-generated questions are starting points, adapt them to what you hear in the first two minutes.
5. Trigger events (funding, hiring surges, leadership changes) are the highest-value signals to act on.
6. Your CRM notes plus AI = personalized prep at scale. Feed AI your existing deal context.
7. A consistent pre-call prep routine built around AI cuts preparation time by 50–70% without sacrificing quality.

Building a Prospect Intelligence Brief

Before any discovery call, you need a one-page brief covering four things: who the company is, what's happening with them right now, what pressures their industry is under, and what your contact cares about professionally. AI assembles this in under three minutes. Paste a LinkedIn URL, a company name, or a recent news headline into ChatGPT or Claude and ask for a structured brief. The output gives you talking points, not a wall of text to skim at 8:59 AM.

The brief becomes most powerful when you layer in trigger events, the specific moments that make a prospect likely to buy. A new CFO typically reviews vendor contracts within 90 days of starting. A Series B announcement means a company is scaling and probably hiring. A competitor's product launch creates anxiety. AI helps you spot these signals quickly when you feed it recent news, job postings, or LinkedIn activity and ask it to identify what's changing for this company.

• Company overview: industry, size, business model, key revenue streams
• Recent news: funding, acquisitions, leadership changes, product launches, layoffs
• Industry pressures: macro trends, regulatory shifts, competitive dynamics
• Contact profile: role, likely KPIs, typical pain points for this title
• Trigger events: anything that happened in the last 90 days that creates urgency
• Competitor context: which alternatives they might be evaluating

Generating Discovery Questions That Actually Work

Generic discovery questions get generic answers. 'What are your biggest challenges?' is a conversation ender dressed as an opener. AI generates specific, role-aware questions when you give it the right inputs. The output should feel like something a senior rep who's worked the account for two weeks would ask, not something lifted from a sales methodology PDF. Specificity in your prompt produces specificity in the questions.

Structure your AI-generated questions across three layers: situational (what's true today), implication (what does that cost them), and vision (what does good look like). This mirrors proven discovery frameworks without requiring you to memorize them. Ask AI to generate five questions per layer for your specific prospect and product. Then trim to the eight or ten that fit the call flow you're planning. You'll never run out of smart things to ask.

1. Situational: 'Walk me through how your team currently handles [process].'
2. Situational: 'How many people are involved in [decision or workflow]?'
3. Implication: 'What happens downstream when [problem] occurs?'
4. Implication: 'How much time does your team spend working around this?'
5. Implication: 'Has this affected any deals or customer relationships?'
6. Vision: 'If you could change one thing about how this works today, what would it be?'
7. Vision: 'What would success look like for you six months after solving this?'

Avoiding Over-Reliance on AI in Discovery

AI prep makes you sharper, it doesn't replace listening. The reps who get into trouble are the ones who walk in with twelve AI-generated questions and treat the call like an interview instead of a conversation. Your prep brief is a hypothesis to be tested, not a script to be executed. The moment your prospect says something unexpected, your job is to follow that thread, not return to question seven on your list.

There's also a data quality risk. AI works from public information, which means it can be months out of date. A company's LinkedIn headcount, a news article about a product launch, a funding announcement, these are starting points. Confirm key facts early in the call. Say: 'I saw you recently [trigger event], how has that changed things for your team?' That question validates your research and shows you did your homework, without betting the call on information that might be stale.

Quick-Reference Cheat Sheet

• Always open your AI prompt with your product, buyer role, and company context.
• Use trigger events (funding, hiring, leadership changes) as your sharpest discovery hooks.
• Generate questions across three layers: situational, implication, vision.
• Tailor questions to the buyer's role and likely KPIs, generic questions get generic answers.
• Treat AI output as a hypothesis, not a script, follow what you hear in the call.
• Verify key facts early in the call rather than assuming AI's public-data research is current.
• Use the hypothesis opener: 'Companies like yours often struggle with X, is that true for you?'
• Review your brief before the call, then put it away and listen.
• After the call, paste your notes into AI and ask it to summarize next steps and flag unanswered questions.

Key Takeaways

• AI compresses pre-call research from 45 minutes to under 5, without sacrificing depth.
• A structured prompt (your product + their role + their company + any recent context) produces specific, usable output.
• Trigger events are the highest-leverage input for discovery, they create urgency and personalization simultaneously.
• Role-specific question frameworks (situational, implication, vision) give AI the structure it needs to generate questions that sound like a senior rep, not a survey.
• The risk isn't using AI too much, it's using it as a crutch. Prep with it, then listen in the call.
• Consistency is the compounding advantage: reps who use this system on every call improve faster than reps who wing it.