Nozz Profile picture
Aug 24 โ€ข 20 tweets โ€ข 3 min read โ€ข Read on X
Sales calls are a fucking goldmine that 99% of people throw in the trash.

Record them. Transcribe them. Feed them to GPT.

Here's how I close deals without being a "sales guy":๐Ÿงต
First, shoutout to @SergeGatari for this game-changer:

Never say "AI" on sales calls.

Say "automation" or "system" or literally anything else.

Tested this all week. People's guard drops INSTANTLY.
Here's what actually happens when you say "AI solution":

Client brain: "Oh great, another tech bro trying to sell me ChatGPT"

When you say "automation system":

Client brain: "This person might actually solve my problems"
The real magic happens AFTER the call.

While other agencies forget 80% of what was discussed...

I have every. single. word.

Names. Pain points. Objections. Budget hints. Everything.
My dead simple process:

- Record with Fathom/Fireflies (free tier works)
- Download transcript
- Drop into Claude/GPT
- Ask: "What follow-up email would close this deal?"

Done.
But here's where it gets disgusting...

GPT remembers shit you don't even notice:

"When you mentioned John from accounting..." "Regarding your Q3 deadline you briefly touched on..." "I noticed Sarah from marketing seemed concerned about..."
Clients think you're a fucking genius.

"Wow, this person actually LISTENS"

Meanwhile you were zoning out thinking about what to eat for lunch.

The transcript caught everything.
Real example from last week:

Client mentioned their CMO "Jessica" once in passing.
I forgot her name 5 seconds later.

My follow-up email referenced Jessica's concerns about implementation timeline.

Client response: "This is exactly what we needed to hear"
The psychology is beautiful:

People drop their guard when they think you truly understand them.

Nothing says "I understand" like referencing specific details they mentioned.

Even if you were barely paying attention.
Here's my exact prompt:

"Based on this sales call transcript, write a follow-up email that:

- References specific pain points discussed
- Names the actual people mentioned
- Addresses their unstated objections
- Moves toward closing the deal"
Pro tip: Clean up the transcript first

Remove your "ums" and "uhs" Fix any transcription errors Delete the small talk

GPT works better with clean data.
The follow-up email framework it creates:

Opening: Reference specific conversation point Problem: Their exact words, not yours Solution: Positioned as "system" not "AI" Proof: Relevant to

THEIR industry CTA: Soft close based on their buying signals
Another psychological hack:

People have fragile egos about their problems.

Never say "our system is better than your team"

Say "our system enhances what your team already does well"

Same outcome. Different feeling.
What kills deals:
"Our AI will replace..."
"Your current process is broken..."
"You're doing it wrong..."

What closes deals:
"This automates the repetitive parts..."
"Enhances your existing workflow..."
"Builds on what's working..."
The transcript also reveals their REAL objections.

What they say: "We need to think about it"

What the transcript shows:
They mentioned budget constraints 3 times, integration concerns twice, and kept asking about support.

Now you know what to address.
My close rate went from 20% to 45% doing this.

Not because I became better at sales.

Because I stopped forgetting important shit.

The transcript remembers everything. GPT connects the dots.
Side benefit:

You can analyse patterns across all your calls.

"What objections come up most?"
"What features do people actually care about?"
"Where do deals die?"

Feed 10 transcripts to GPT. Get a masterclass in your own sales process.
Tools I use:

Recording: Fathom (free)
Transcription: Built into Fathom or FireFlies
Analysis: Claude for nuance, GPT for structure
CRM: Notion (with transcripts attached)

Total cost: $0-20/month
The bottom line:

I'm shit at sales. Can't remember names. Zone out constantly.

But I close deals because:

- Everything gets recorded
- GPT catches what I miss
- Follow-ups reference exact pain points
- Clients feel truly heard
Stop trying to be a better salesperson.

Start being a better listener.

Or just record everything and let GPT do the listening for you.

Your commission checks won't know the difference.

โ€ข โ€ข โ€ข

Missing some Tweet in this thread? You can try to force a refresh
ใ€€

Keep Current with Nozz

Nozz Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @NoahEpstein_

Aug 24
Claude just killed the $500/month AI agent industry with one update.

Got DMs asking how I replaced $6K worth of AI tools with free connections.

Here's how to turn Claude into your business's AI brain: ๐Ÿงต
First, let me show you what just happened...

OpenAI: "Pay $200/month for o1"
Perplexity: "Pay $20/month for search"
: "Pay $100/month for automation"
Zapier: "Pay $300/month for connections"

Claude: "Here's MCP. Connect everything. Pay nothing extra."

Game fucking over.Make.com
MCP = Model Context Protocol

Fancy words for: Claude can now directly control your entire tech stack.

No APIs.
No webhooks.
No complex integrations.

Just natural language commanding your business tools like a puppet master.
Read 14 tweets
Aug 22
99% of AI builders are competing on features while smart operators are selling $47K "chatbots" to lawyers who can't even PDF a document.

The real AI goldmine isn't in the tech - it's in the intelligence gap.

A thread on how people are printing money from digital stupidity ๐Ÿงต
Just analysed 500+ automation projects from the last 6 months.

The pattern is insane:

Companies are sitting on MILLION DOLLAR problems while asking for $2K solutions.

They literally don't know what they don't know.
What businesses think AI does:

โŒ Chatbots that break after 3 questions
โŒ Email templates with {firstname}
โŒ Copying data from spreadsheet A to B
โŒ "Hey Siri but for our company"

Their imagination is stuck in 2019.
Read 15 tweets
Aug 20
Retards think automation is about saving time.

Gs know it's about psychological arbitrage.

I studied 73 viral n8n builders making $50K+/month.

Here's the message they all send that template peasants miss: ๐Ÿงต
1. They're slightly unhinged

Retards: "Let me automate my email responses"

Gs: "Let me build a system that stalks my competitors' every move, analyse their
weaknesses, and automatically undercuts their pricing by 7.3%"

The best automations are borderline creepy.
2. They see money where others see tasks

Retards automate TO-DO lists.

Gs automate ATM machines.

Example: One builder automated Reddit complaint scraping.

Now charges $3K/month to brands for "reputation monitoring."

Same tech. Different vision.
Read 10 tweets
Aug 18
OpenAI just admitted AI agents will replace 92 million jobs by 2030.

But they buried the real story:

170 million NEW jobs are being created.

I spent 6 months researching what separates the winners from the losers.

Here are the 7 "recession-proof" skills nobody's teaching you: ๐Ÿงต
First, let's address the elephant in the room.

The job apocalypse is real:
- 77.4% of companies already use AI
- Entry-level positions vanishing daily
- Microsoft: 30% code is AI-written, then fired 40% of engineers
- IBM replaced entire HR departments

But that's only half the story...
Here's what the doomers won't tell you:

For every job AI kills, it creates 1.8 new ones.

The problem?
These new jobs require skills that didn't exist 2 years ago.

Skills your college didn't teach.
Skills bootcamps don't cover.
Skills that sound made up.

Until you see the salaries...
Read 24 tweets
Aug 17
AI can now predict human behaviour better than psychologists.

Not in 10 years. Not in 5 years. Right NOW.

I just discovered the psychological principles that let you manipulate AI to message directly to people's subconscious.

A thread on the dark psychology of artificial intelligence: ๐Ÿงต
First, the bombshell nobody's talking about:

Researchers fed an AI model data from 160 psychology studies.

10 million human decisions.
60,000 participants.

Result? It predicts human behaviour with 95% accuracy.

We've created a mind reader.
But here's where it gets terrifying...

While AI learns to predict us, we're learning to predict IT.

The same psychological triggers that work on humans?
They work on AI too.

I've been testing this for 6 months.
The results will disturb you.
Read 14 tweets
Aug 13
I built the same workflow in Google Opal and n8n.

One took 5 minutes. One took 20.

Google Opal VS n8n

(Full comparison + when to use each) ๐Ÿ‘‡ Image
Google Opal (Launched July 24, 2025):

โœ“ FREE (for now)
โœ“ No-code visual builder
โœ“ US-only beta
โœ“ Build mini-apps in minutes
โœ— Can't self-host
โœ— Limited to Google's ecosystem
โœ— No complex workflows
โœ— Can't connect to most business tools
n8n:

โœ“ Open source (own your data)
โœ“ 400+ integrations
โœ“ Self-hostable
โœ“ Complex multi-step workflows
โœ“ Works globally
โœ— Steeper learning curve
โœ— Requires technical setup
โœ— Costs money to host
Read 7 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

:(