How to Write a Sales Prospect Brief with AI in 2026

A strong prospect brief does three things: it shows the seller has done real research on the account (not just pulled the LinkedIn headshot), it identifies the specific pain or initiative this prospect is likely working on, and it gives the seller a hypothesis to test in the first 10 minutes of the discovery call. The brief that turns a cold meeting into a real conversation isn't the longest — it's the one that surfaces the one or two facts the seller can mention to make the prospect lean in. AI is excellent at producing the structural and synthesis layer of a brief in three minutes. The actual research — earnings call quotes, recent press, LinkedIn signals, mutual connections — is yours to provide.

This is a practical walkthrough for writing a prospect brief with AI that earns the prospect's attention.

What a winning prospect brief contains

Before you can use AI well, you need to know what good looks like:

• Header block — prospect name, title, company, account stage (new logo / expansion / renewal), call date and type
• Account snapshot — company size, industry, recent funding or strategic events, tech stack signals if known
• Prospect context — role, tenure in role, prior companies, any public signals (conference talks, podcasts, posts) that suggest what they care about
• Likely initiatives — the 2-3 things this person is probably working on right now (based on role + company stage + recent news)
• Pain hypothesis — the specific problem we hypothesize they're solving for; the evidence behind that hypothesis
• Connection signal — mutual LinkedIn connections, shared employers, shared cities, shared customers in their space
• Discovery questions — 3-5 questions designed to test the pain hypothesis without leading
• What to avoid in the call — topics that misread the prospect (e.g., pitching a feature they already have, talking pricing too early)
• Next-step ask — what we want at the end of this call

Sellers who consistently book follow-up calls are the ones whose first call surfaces a specific, validated pain. AI handles the structural and synthesis layer; you provide the research inputs (the prospect's actual posts, the company's recent press, the seller's own knowledge of the market).

The right prompt structure

The mistake most sellers make on first try is asking AI for "research on Acme Corp" with no context. The prompt that actually works gives the AI the research the seller has already gathered and asks for synthesis:

<task>Write a sales prospect brief for a discovery call.</task>

<context>
Seller: [SELLER NAME], Account Executive
Product context: We sell a procurement-automation platform; target buyer
is VP/Director of Procurement at companies $500M-$5B revenue.

Prospect: Maria Chen, VP of Procurement
Company: Acme Manufacturing, ~$1.2B revenue, industrial components
Account stage: New logo; cold outbound that converted to a discovery call

Prospect background (from research):
- 6 years at Acme; promoted to VP 14 months ago
- Prior: 8 years at Globex (similar industry), led procurement
  transformation per her LinkedIn bio
- Recent LinkedIn post (3 weeks ago) — celebrating the team for
  "completing supplier consolidation Phase 1" — implies multi-phase project
- Spoke at ISM 2025 on "centralizing tail spend"

Company signals:
- Q1 2026 earnings call — CFO mentioned "ongoing supply chain
  efficiency initiatives" and called out "indirect spend visibility"
  as a focus area
- Hired a Director of Strategic Sourcing in March 2026
- Recent 10-K: 1,200+ active suppliers, ~$340M annual indirect spend

Mutual connections (LinkedIn):
- Our customer at Stark Industries (Procurement Director) — open
  to a warm intro if asked

Call: 30 minutes, May 23, 2026, 10am PT
Goal: Validate pain hypothesis; secure a follow-up scoping conversation
</context>

<instructions>
- Structure: account snapshot, prospect context, likely initiatives,
  pain hypothesis (with evidence), connection signal, 4 discovery
  questions, what to avoid, next-step ask
- Tone: tactical, peer-to-peer, not breathless
- Use the actual research — quote the LinkedIn post excerpt and the
  CFO earnings call language where it sharpens the hypothesis
- Pain hypothesis must be specific and tied to the evidence
- Discovery questions: open-ended, not leading, designed to test
  (not confirm) the hypothesis
- 500 words maximum
</instructions>

<avoid>
- Inventing company facts, headcount, financials, or quotes not in context
- Inventing the prospect's prior employers, tenure, or posts
- Pitching the product before validating the pain
- Generic discovery questions ("what keeps you up at night")
- Treating the LinkedIn bio as the whole research story
- Claims about competitors the prospect uses without evidence
</avoid>

The structure: the research already gathered, the product context, the call goal, and explicit instructions about what NOT to invent. The AI synthesizes the brief; you provide the verified facts.

What to never let AI do

Invent facts about the prospect or company. AI will produce plausible-sounding company metrics, prospect tenure, and quotes if you don't constrain it. Every fact in the brief should come from a source the seller can name. If you can't source it, leave it out.

Pull from training-data knowledge of the company. AI's training data may be 12+ months stale and may have absorbed inaccuracies. Use only the research the seller has gathered fresh. Don't ask the AI "what do you know about Acme Manufacturing."

Generate discovery questions that lead the prospect. "How important is procurement automation to your strategy?" is a leading question that primes the prospect for a yes. Open questions ("walk me through the supplier consolidation work — what's working and what's the next bottleneck") test the hypothesis instead of confirming it.

Pitch instead of qualify on the first call. AI may default to "and here's how our product solves that" — that's the wrong move on a discovery call. The brief should support qualification, not pitch.

Skip the next-step ask. A discovery call without a defined next-step ask ends with "we'll follow up." The brief should specify what the seller is going for at the end of the call (e.g., scoping conversation with the technical buyer; data-gap workshop; pricing conversation in week 3).

Common mistakes

Generic pain hypothesis. "They probably want to save money on procurement" is true for every procurement buyer. "Phase 1 was supplier consolidation; Phase 2 is likely visibility into the consolidated supplier set — and the CFO's emphasis on indirect spend visibility on the earnings call supports this" is a specific, testable hypothesis.

Research that doesn't shape the call. A brief that lists 10 facts but doesn't tie any to a discovery question is research without strategy. Every fact in the brief should connect to a question, an opening line, or a thing to avoid.

Ignoring the connection signal. Mutual connections, shared customers, or referenceable peers in the prospect's space are some of the highest-trust signals in the first call. Surface them in the brief.

Wrong-stage discovery questions. Asking pricing questions on a discovery call alienates the prospect. Asking technical-architecture questions on a first call with a business buyer wastes time. Match the questions to the prospect's role and the call stage.

No "what to avoid" section. Most sellers know what to ask. Fewer know what to avoid. A brief that flags "don't lead with the AI/automation angle — her last post emphasized people-led process change" prevents a bad first impression.

What to never put in a prospect brief without consideration

• Personal information about the prospect not relevant to the business context (family, health, personal politics)
• Speculation about the prospect's compensation, performance, or job security
• Information sourced from non-public channels without authorization
• Competitor displacement claims without evidence
• Pricing or commercial terms (these belong in a different document at a different stage)

These aren't AI-specific risks — they apply to any prospect research artifact. AI can produce them quickly if you don't constrain; the seller's review step catches them.

The free tool that handles this for you

If you don't want to engineer the prompt every time, the Prospect Research Brief Generator on AI Career Lab is pre-configured for the structure that turns cold meetings into real conversations. It produces briefs with the elements above, in the tactical peer-to-peer tone that respects the seller's time.

Pair it with the Sales Follow-Up Generator for the post-call note that books the next meeting, the Sales Call Prep Generator for late-stage call planning, and the Sales Proposal Generator for the proposal that lands the close.

Free with an AI Career Lab account, capped at five runs per day on the free tier.

Try it on your next discovery call

Pick the next discovery call on your calendar. Spend 15 minutes pulling the actual research — recent LinkedIn posts, earnings call language if public company, mutual connections, press from the last 90 days. Run the inputs through the tool above. Compare to the brief you'd write by hand — note how much more specific the pain hypothesis reads when it's tied to evidence.

Create your free AI Career Lab account and try the sales tools today. No credit card.

---

This article is general guidance for sales professionals. AI-generated prospect briefs are starting drafts requiring seller review for factual accuracy. Use only research the seller has gathered from public or properly-licensed sources. Respect prospect privacy and applicable regulations (GDPR, CCPA, professional codes of conduct).