Prompting

In partnership with

You open ChatGPT and type "write me a cold email."

You get back generic garbage that sounds like every other desperate sales rep on LinkedIn. In this email I’ll provide you with the exact prompt framework you should be using every time you open your LLM of choice (not just ChatGPT)

TL;DR: I've built 25 sales-specific prompts using this framework - covering cold outreach, discovery prep, objection handling, proposal generation, and more.

Grab them here 👉 https://start.snowballn.com/salespromptblueprint/

Unlock ChatGPT’s Full Power at Work

ChatGPT is transforming productivity, but most teams miss its true potential. Subscribe to Mindstream for free and access 5 expert-built resources packed with prompts, workflows, and practical strategies for 2025.

Whether you're crafting content, managing projects, or automating work, this kit helps you save time and get better results every week.

Get Your Free Resource Kit

You spend 30 minutes editing it until it's usable.

(it would’ve been faster just to write the email yourself)

Then you wonder why everyone says AI is the future…

Here’s the thing:

The AI isn't broken. Your prompt is.

Most sellers treat AI like a search engine. They give vague instructions and expect magic.

That's not how this works.

Prompt engineering is about talking to AI like you'd brief another colleague (a junior one at that).

Be specific. Give context. Set constraints. Explain why things matter.

Do that, and you go from spending 30 minutes cleaning up garbage to getting publication-ready output in 30 seconds.

What changes when you get this right:

→ You stop wasting hours on emails, account plans, and competitive research
→ Your AI output stops sounding generic and starts sounding like you
→ Your whole team can use the same templates and get consistent results
→ You build systems that scale instead of grinding harder

The prompt framework

Here's how every high-performing prompt works.

Seven components.

Use them all.

For every prompt.

1. Role

Tell the AI who it should act as.

This sets perspective, expertise level, and how it makes decisions.

Example: "Act as a VP of Sales who has built outbound teams at three successful SaaS companies..."

2. Task

The specific action you want done.

Be crystal clear about the deliverable.

Example: "Draft a cold outreach sequence for enterprise CFOs..."

3. Format

How you want the output structured.

This kills the back-and-forth revision dance.

Example: "Present as 3 emails: initial outreach, +3 day follow-up, +7 day breakup email"

4. Warning

What to avoid. What not to do.

This keeps the AI from going sideways.

Example: "No generic value propositions. No buzzwords like 'synergy' or 'game-changing.' No emails longer than 100 words."

5. Reasoning

Why certain choices matter. What trade-offs exist.

Example: "Each email should address a specific business pain. Balance urgency with authenticity. We want to break through inbox noise without sounding desperate."

6. Stop Conditions

When is the task actually done?

Be explicit.

Example: "Stop once you've delivered 3 emails, each with a subject line, and included A/B test variants for the first email subject."

7. Context Dump

All the background info that matters.

More context = better output. Always.

Example: "Our ICP: Mid-market B2B SaaS companies ($5M-$50M revenue) struggling with manual sales ops. Our product automates pipeline reporting. Average deal size: $25K. Sales cycle: 45 days."

It's 7 parts. Learn it once, use it forever.

Role. Task. Format. Warning. Reasoning. Stop Conditions. Context Dump.

Each one tells the AI exactly what to do, how to do it, and when to stop.

I've built 25 sales-specific prompts using this framework - covering cold outreach, discovery prep, objection handling, proposal generation, and more.

Grab them here 👉 https://start.snowballn.com/salespromptblueprint/

Power to you,

Mike