AI for Sales & GTM
I built and sold an AI company (Lyne.ai). Worked with 200+ B2B teams. Spent two years testing every AI tool that claimed to help with sales. 95% of it is hype. This guide is about the 5% that works.
Not the Hype
The real state of AI in sales
Every AI company claims their tool will "10x your outbound" and "replace your SDR team." Both are lies.
The companies winning with AI aren't replacing humans. They're augmenting them. At Lyne.ai, we learned this the hard way. Our AI could write decent emails, but the best performing campaigns still needed human oversight. The magic happened when AI handled the research and humans handled the strategy.
✓ What AI does well in sales
- Research and data analysis
- Content personalization at scale
- Workflow automation
- Call transcription and analysis
✗ What AI sucks at (still)
- Building genuine relationships
- Complex buying processes
- Handling objections with empathy
- Closing deals
My AI Stack (What I Actually Use)
No AI SDRs. No AI closers. Just AI doing what it does best so humans can do what they do best.
Where AI Shines
Using Claude for prospecting research
This is where AI shines. What used to take 15 minutes per prospect now takes 30 seconds.
My Research Prompt (Claude)
// Paste this into Claude with a company name
I'm researching [COMPANY] for a potential partnership.
Please analyze:
1. Recent funding, acquisitions, or major announcements
2. Leadership changes or key hires in the last 6 months
3. Their current GTM strategy based on job postings
4. Potential pain points or growth challenges
5. Best angle for outreach based on the above
Format as a brief summary with bullet points.
Focus on actionable insights.
Takes Claude 30 seconds. Would take 15 minutes manually.
Advanced Research Workflow
Use Clay to pull company data (domain, industry, size)
Feed this to Claude for deeper research
Cross-reference with intent data (6sense, Bombora)
Generate personalized talking points
Create custom email sequences
This isn't theoretical. I do this for every new client prospect.
The Right Way
AI email writing that doesn't sound like AI
Most AI-generated emails are garbage. Too formal, too long, too obvious. But with the right prompts, AI can write emails that sound human.
My Email Writing Prompt (Claude)
Write a cold email for [PROSPECT] at [COMPANY].
Context:
- Their role: [TITLE]
- Company info: [BRIEF DESCRIPTION]
- Recent trigger event: [NEWS/FUNDING/HIRE]
- My value prop: [SPECIFIC BENEFIT]
Style guidelines:
- Casual, conversational tone
- 50-75 words maximum
- No corporate speak or buzzwords
- Reference the trigger event naturally
- Soft ask, not hard sell
- End with a simple question
Write 3 variations.
The Secret
Specific context + clear constraints = better output. Most people give AI vague prompts like "write a sales email." Be specific about tone, length, and purpose.
✓ AI Email Do's
- Always provide specific context
- Set clear word limits
- Ask for multiple variations
- Edit before sending (always)
✗ AI Email Don'ts
- Don't use it for follow-ups (too generic)
- Don't let it write your entire sequence
- Don't skip human review
- Don't use the same prompt for everyone
Honest Review
AI SDR tools: what works, what doesn't
I've tested every AI SDR tool: Artisan, Smartlead AI, Reply.io AI, Outreach AI, and dozens of smaller ones.
The Promise: AI that can research prospects, write personalized emails, handle objections, and book meetings automatically.
The Reality: Most are glorified email generators with fancy marketing.
Artisan Ava (AI SDR)
Good
Decent research capabilities, integration with major data sources
Bad
Emails still sound robotic, can't handle complex responses
Verdict
Useful for high-volume, low-touch campaigns only
Smartlead AI Features
Good
Email optimization, send time prediction
Bad
AI writing is generic, lacks industry knowledge
Verdict
Use Smartlead for sending, not for AI
Reply.io AI
Good
Email performance analysis, A/B testing
Bad
AI sequences are cookie-cutter
Verdict
Good analytics, skip the AI writing
My Recommendation
Use AI for research and first draft writing. Keep humans in the loop for everything else. AI objection handling, meeting scheduling, follow-up sequences, and prospect scoring all still need human judgment.
The Good Stuff
Building AI-powered GTM workflows
The real power of AI isn't in replacing humans. It's in automating the tedious stuff so humans can focus on high-value activities.
AI Research Pipeline
Clay finds new prospects matching ICP
Webhook triggers Claude research prompt
AI analyzes company news, funding, hires
Results feed into personalized email templates
Human reviews and approves before sending
Content Personalization at Scale
Clay pulls prospect data (title, company, industry)
AI generates industry-specific case studies
Dynamic email templates insert relevant examples
Human reviews high-value prospects manually
Meeting Prep Automation
Meeting booked → webhook to research pipeline
AI analyzes prospect's LinkedIn, company site, recent news
Generates meeting prep: background, pain points, questions
Recommends relevant case studies to mention
Tools for Building AI Workflows
The Fastest-Growing Role
The GTM Engineer role
This is the fastest-growing role in B2B. Companies are realizing they need someone who understands both GTM strategy and technical implementation.
The average B2B company uses 20+ GTM tools. Most are poorly integrated. GTM Engineers fix this by building unified workflows that span the entire customer journey.
Key Responsibilities
- Build data pipelines (Clay, Zapier, APIs)
- Automate repetitive GTM tasks
- Integrate tools across the stack
- Custom dashboards and reports
- Implement AI workflows
Required Skills
- No-code/low-code tools (Clay, Zapier)
- Basic API knowledge
- Data analysis and visualization
- Understanding of GTM processes
- Project management
GTM Engineer Salary Range
This role didn't exist 3 years ago.
| Level | Salary Range |
|---|---|
| Entry level | $80,000-$120,000 |
| Mid-level | $120,000-$180,000 |
| Senior | $180,000-$250,000+ |
How to Become a GTM Engineer
Master no-code tools (start with Clay and Zapier)
Learn basic API usage and webhooks
Understand the entire GTM stack
Build portfolio projects demonstrating automation
Get certified in relevant tools
I'm seeing 2-3 GTM Engineer job postings per week in my network. It's the future of GTM operations.
Questions
FAQ
Can AI replace SDRs and sales reps?
No. AI excels at research, data analysis, and content generation, but humans are still needed for relationship building, complex objection handling, and closing deals. The best results come from AI + humans working together.
What AI tools actually work for sales?
Claude and ChatGPT for prospecting research and email writing. Gong for call analysis. Most "AI SDR" tools are overhyped and underperform compared to humans with AI assistance.
How do I write AI emails that don't sound robotic?
Use specific context, set clear style guidelines (casual tone, word limits), ask for multiple variations, and always edit before sending. Generic prompts produce generic emails.
What is a GTM Engineer?
Someone who builds and maintains the technology infrastructure for GTM activities — data pipelines, workflow automation, tool integration, AI implementation. Salaries range from $80k-$250k+.
Master Claude for GTM
Learn to build automated outbound systems using Claude AI. Real prompts, workflows, and playbooks that are driving results right now.