--- title: "Call prep, rehearsal and post-call follow-up" description: "Better calls start with better questions Discovery calls decide deals. AI can help you prepare in 15 minutes, rehearse the hard moments, and turn the…" url: https://optimizeall.com/learn/prompt-engineering-for-content-and-sales/call-prep-and-discovery-prompts updated: 2026-10-05 --- Prompt Engineering for Content & Sales · Prompts for sales: outreach, calls, follow-ups and proposals · lesson 9 of 15 · 13 min # Call prep, rehearsal and post-call follow-up ## Better calls start with better questions Discovery calls decide deals. AI can help you prepare in 15 minutes, rehearse the hard moments, and turn the call into accurate notes and next steps. What it can't do is listen for you: the call itself is where your judgment matters most. ## The 15-minute call-prep pack ```text 1. CALL BRIEF I have a [30]-minute discovery call with [role] at [company] about [topic]. From this account brief and CRM notes (anonymized where needed): [paste] Give me: what we know, what we don't know, the 3 most likely priorities (as hypotheses), and the single outcome I should aim for by the end of the call. ``` ```text 2. QUESTION BANK Write 12 discovery questions in 4 groups: situation, problem, impact, decision process. Open-ended, one question at a time, no leading questions, plain language. Mark the 5 I must ask if time runs short. ``` ```text 3. OBJECTION PREVIEW List the 5 objections this buyer is most likely to raise, with an honest 2-sentence response for each using only our offer sheet and approved proof. Mark any response that needs data I haven't given you. ``` ```text 4. ROLE-PLAY (voice mode) You are [role] at a [company type]. You're busy and skeptical; you've been burned by a similar tool before. I'm running discovery for [offer]. Answer realistically, push back twice, and don't volunteer information unless I ask good questions. After 10 minutes, stop and score my questions: which uncovered real information, which were leading, what I missed. ``` ## During the call - Use your question bank as a guide, not a script. - Take rough notes on **their words** for pains and priorities; you'll reuse them. - If you use meeting recording or transcription, tell participants and follow your company policy and local law. ## After the call: notes to next steps ```text Here are my notes / the transcript from a discovery call with [company]: [paste]. Produce: 1. Their top 3 priorities in their own words (quote them) 2. Current situation and what they've tried 3. Impact if nothing changes (only what they said) 4. Decision process: who's involved, timeline, budget signals (write "Unknown" if not stated) 5. Agreed next step with date and owner 6. CRM summary in 5 lines 7. A follow-up email draft (under 150 words) reflecting their words, with one next step and two time options Finally: list any promise, number or claim in the email that I should confirm before sending. ``` ## Worked example (illustrative) James sells HR software to mid-size companies in the UK. Before a call with a retail chain's HR director, he runs the prep pack. The question bank includes "Walk me through what happens when a store manager needs cover for a sick shift." The voice role-play (with the AI playing a skeptical HR director) shows he leads with product features too early; he moves them to the end. On the call, the director says, "Our managers spend Sunday nights on WhatsApp finding cover." The post-call prompt puts that quote at the top of the follow-up, and the "confirm" list flags an implementation timeline he mentioned loosely; he checks it with his implementation team before sending. ## Hands-on: prep your next real call Run prompts 1 to 3 the day before, prompt 4 for 10 minutes in voice mode, and the post-call prompt within an hour of the call. **Before:** you arrive with the prospect's website open and a mental list of product features. **After:** you arrive with the outcome you want, five must-ask questions, honest answers to likely objections, one rehearsal behind you, and a follow-up that goes out within the hour quoting the buyer's own words. ## Using connected assistants and CRM AI If your assistant is connected to your email, calendar or CRM (where your organization allows), it can assemble the call brief for you: past emails with the account, the last meeting's notes and open opportunities. Many CRMs and meeting tools now include their own AI call summaries. Treat these like any other draft: check that quotes are accurate, that "decisions" were really agreed, and that nothing sensitive is written into shared records. Keep a human review between any AI summary and the CRM fields your forecast depends on. ## Pitfalls - Reading questions like a survey; listen and follow up. - Treating the role-play as a real buyer's opinion; it's practice. - Letting the AI fill "Unknown" fields in your CRM with guesses. - Recording without consent. ## How to measure success Track the share of calls that end with a concrete next step, how quickly follow-ups go out, and stage conversion from discovery to proposal. ## Video lecture: Call prep, rehearsal and post-call follow-up Lecture coming soon · 11 chapters · about 8 minutes. Read the full transcript below. 1. Call prep and discovery 2. Why discovery decides deals 3. The journalist analogy 4. The prep pack 5. Rehearse in voice mode 6. After the call 7. Example 1: the must-ask five 8. Example 2: the Sunday-night WhatsApp quote 9. Watch me do it 10. Common mistakes 11. Recap and try this now ## Lecture transcript ### Call prep and discovery Think about the best discovery call you've ever had. You asked a question, the buyer paused, and then told you something they'd never told a salesperson before. That moment doesn't happen by accident. It happens because you prepared the right questions and had the confidence to ask them. In this lecture you'll learn a fifteen minute call prep pack, how to rehearse in voice mode with a skeptical AI buyer, and how to turn call notes into accurate CRM entries and a follow up within the hour. You'll see a simple example and a UK HR software call, and watch me prep a real call. ### Why discovery decides deals Why does discovery matter so much? Because deals are won or lost on understanding. The salesperson who understands the buyer's situation, the impact of the problem and how the decision gets made can shape a proposal that fits. There's a second benefit people miss. The buyer's own words, their description of the pain, become your best sales copy, in the follow up, the proposal, and even your marketing. And third, speed after the call builds trust. A clear follow up within the hour, quoting what they said, signals that you listened. AI helps with all three, as long as you do the listening. ### The journalist analogy Think like a good journalist. Before an interview, they research the person and the organization. They prepare open questions, but they don't read them like a survey: they listen and follow up. They get permission before recording. And afterward, they quote accurately, never putting words in someone's mouth. Discovery works the same way. AI can be your research assistant and your rehearsal partner, and it can help you write up the story afterward. But the interview itself, the listening and the follow up questions, is yours. ### The prep pack Here's the prep pack. Prompt one, the call brief: from the account brief and your CRM notes, what we know, what we don't, the three most likely priorities labeled as hypotheses, and the single outcome to aim for by the end of the call. Prompt two, the question bank: twelve discovery questions in four groups, situation, problem, impact and decision process. Open ended, no leading questions, plain language, with the five must asks marked. Prompt three, the objection preview: five likely objections, each with an honest two sentence response using only your offer sheet and approved proof, and a flag where you need data. Fifteen minutes, and you walk in prepared. ### Rehearse in voice mode Prompt four is rehearsal, and it's where confidence comes from. In voice mode, set the frame. You're this role at this kind of company. You're busy and skeptical, and you've been burned by a similar tool before. Answer realistically, push back twice, and don't volunteer information unless I ask good questions. After ten minutes, stop and score my questions: which uncovered real information, which were leading, and what I missed. It feels awkward the first time. By the third time, you'll notice your questions getting sharper. Just remember: it's practice. The AI buyer is an average, not your real buyer. ### After the call After the call, within the hour, paste your notes or the transcript, recorded with consent, into the post call prompt. Ask for their top three priorities in their own words, quoted. Their current situation and what they've tried. The impact if nothing changes, only what they actually said. The decision process: who's involved, timeline, budget signals, with unknown where it wasn't discussed. The agreed next step. A five line CRM summary. And a follow up email that reflects their words with one next step and two time options. Then the safety net: list any promise, number or claim in the email I should confirm before sending. ### Example 1: the must-ask five A simple example of why the must ask five matter. A sales rep's thirty minute call starts fifteen minutes late because the buyer's previous meeting overran. Without preparation, he'd waste the time on small talk and a product tour. Instead, he goes straight to his five starred questions: what's prompting you to look at this now, what happens today when this goes wrong, what's that costing you, who else will weigh in on the decision, and what would a good next step look like for you? In fifteen minutes he learns the trigger, the impact, the decision makers and the next step. That's a successful discovery call, even when half of it disappeared. ### Example 2: the Sunday-night WhatsApp quote Now a fuller example, with illustrative details. James sells HR software to mid size companies in the UK. Before a call with a retail chain's HR director, he runs the prep pack. His voice role play shows he jumps to product features too early, so he moves them to the end. On the call, one question lands: walk me through what happens when a store manager needs cover for a sick shift. The director says: our managers spend Sunday nights on WhatsApp finding cover. James writes that down word for word. The post call prompt puts that quote at the top of the follow up. And the confirm list flags an implementation timeline he'd mentioned loosely. He checks it with his implementation team before sending. ### Watch me do it Let me prep a real style call. It's a discovery call with the operations director of a restaurant group in Dubai, about our inventory software. Prompt one: the brief gives three hypotheses, like food waste and supplier price changes, and an outcome to aim for: agree a site visit. Prompt two: twelve questions, and I star five. Prompt three: the likely objections, including we already use spreadsheets, and an honest response. Then ten minutes of voice rehearsal. The AI director pushes back on price and asks about integration with their point of sale. The scorecard says two of my questions were leading. I rephrase them. After the real call, I paste my notes into the post call prompt. The summary quotes her top priority and marks budget as unknown. I don't guess it. That's next call's question. ### Common mistakes The common mistakes. Reading your questions like a survey. The question bank is a guide, so listen and follow up on what matters. Treating the role play as a prediction of what the real buyer will say. It's rehearsal. Letting the AI fill unknown CRM fields with plausible guesses, which quietly corrupts your pipeline data. And recording or transcribing calls without telling participants, which breaks trust and, in many places, the law. Follow your company's policy and ask first. ### Recap and try this now Let's recap. Before the call, run the prep pack: a call brief with a clear outcome, a question bank in four groups with five must asks, and honest objection responses. Rehearse for ten minutes in voice mode with a skeptical AI buyer. During the call, listen, follow up, and capture their words. Afterward, use the post call prompt for priorities in their own words, unknowns marked as unknown, a CRM summary and a follow up within the hour, with every promise confirmed. Here's your try this now. Use the full pack from the lesson text on your very next discovery call, and measure how quickly your follow up goes out. ## Key takeaways - Prepare with a call brief, a question bank (situation, problem, impact, decision), and honest objection responses. - Rehearse in voice mode with a skeptical role-played buyer who scores your questions. - After the call, capture priorities in the buyer's own words, mark unknowns as Unknown, and send a follow-up within the hour. - Get consent before recording, and confirm any promise or number in the follow-up before sending. ## Try it Use the full prep pack and a 10-minute voice rehearsal before your next discovery call, then send the follow-up within an hour. - [Previous: Prospect research and outreach prompts](https://optimizeall.com/learn/prompt-engineering-for-content-and-sales/prospect-research-and-outreach-prompts) - [Next: Templates for sales emails, follow-ups and ad copy](https://optimizeall.com/learn/prompt-engineering-for-content-and-sales/templates-for-sales-and-ads) - [All lessons of Prompt Engineering for Content & Sales](https://optimizeall.com/learn/prompt-engineering-for-content-and-sales)