Podcast Summary: How I AI

Episode: How this CEO turned 25,000 hours of sales calls into a self-learning go-to-market engine

Guest: Matt Britton (CEO of Suzy)
Host: Claire Vo
Date: November 10, 2025

Overview

In this episode, Claire Vo interviews Matt Britton, CEO of Suzy, about building an automated, AI-powered go-to-market (GTM) workflow by leveraging over 25,000 hours of customer and sales call transcripts. The conversation breaks down Matt’s practical approach to automating and enriching sales, marketing, and operational processes—spanning from generating customer summaries to automating Google ad keywords and content creation—using tools like Zapier and AI language models.
Matt demonstrates how a single customer call becomes a multi-faceted operational asset and shares philosophy, technical hacks, and organizational lessons valuable to any business leader embracing AI-driven transformation.

Key Discussion Points & Insights

1. Identifying the Core Problem and Data Asset

• Sales Team’s Pain Point: Sales reps struggled to access key customer insights from past calls and understand how to tailor their approach.
  • “My sales team was consistently telling me that they just didn't know how to find anything. They didn't know how to find what customers were interested in.” (00:00, Matt Britton)
• Realization: Post-pandemic, Suzy had recorded 25,000+ hours of calls using Gong—an untapped goldmine of customer insights.
• Epiphany:
  • “If you think about understanding information about your customers and your business, there's no better source of truth.” (04:32, Matt Britton)

2. Building on No-Code Tools & Personal Agency

• Why Zapier? Matt, not an engineer but a technical founder, embraced Zapier for stitching together multiple tools despite not coding himself. Now, AI integrations within Zapier expand what’s possible.
  • “I've always been fairly technical, but I've never been a coder...I needed to...stitch things together to be more efficient.” (03:17, Matt Britton)
• Learning by Doing: Matt advocates leaders build automations themselves to gain intuition and influence future development.
  • “It is not sufficient to instruct your engineers to build AI...this is a moment for actual hard skill building in leaders.” (09:13, Ariel)
• Analogy:
  • “If you bring your car in and you don't know anything about fixing a car...But if you actually just open up the hood and you understand how transmission works…that’s sort of the same analogy.” (09:35, Matt Britton)

3. Technical Walkthrough: From Call to Automation

Trigger and Data Pipeline (09:35–14:01)

• Trigger & Scraping: Uses Browse AI with Zapier to scrape call transcripts directly from Gong via unique call IDs—even hacking around the lack of an official API.
  • “There’s always a way…just because the tool doesn’t do it doesn’t mean it can’t be done.” (12:32, Matt Britton)
• Downstream Processing: Strips HTML, normalizes text, enriches with Google Sheets lookups to connect transcripts to brands and users.

Orchestration & LLMs (17:16–23:25)

• Automation Structure: Prefers Zapier's sequential (step-based) workflow over more complex "agent" style tools, but sees agents as the next horizon.
• Prompt Engineering:
  • Core summarization prompt extracts company name, participants, call overview, sentiment analysis (scored 1–10), key actions, next steps—across all calls.
  • “Assess the overall customer sentiment...This is the key thing because it allows us to quantify customer sentiment over time and we actually benchmark this against actual churn.” (20:22, Matt Britton)

Multi-Tentacled Output: How One Call Powers Multiple Functions

1. Customer Success & Internal Ops

• Automated Summary Distribution: After each call, an AI-generated summary—including sentiment score, key points, and next steps—is posted to a dedicated Slack channel.
  • “Every time a customer call is done, it just pops up on Slack. And...I'm really able to get a sense of the pulse of the company.” (23:25, Matt Britton)
• Churn Risk: Calls flagged with low sentiment (≤7) trigger alerts in a “churn early warning system,” prompting proactive intervention.
  • “If a customer is not happy…we don't have to ask somebody how that call went...it's just here.” (26:17, Matt Britton)

2. Sales & Coaching

• Real-Time Feedback: After each call, reps receive AI-generated coaching emails highlighting strengths and areas for improvement.
  • “It actually creates a feedback note to the person on the call...Here's what you did right, here's what you did wrong.” (28:34, Matt Britton)
• Follow-up Email Drafts: Automated, customized follow-up emails are sent right after the call—human-in-the-loop ensures final tweaks.
• Performance Analytics: All feedback stored in a dataset for trend analysis and unbiased performance review.

3. Marketing & Demand Generation

• Keyword Extraction & SEM Automation: AI identifies product-interest keywords in transcripts, which are pushed to Google Ads campaigns.
  • “Analyze the key areas of interest data...output a bunch of keywords that we should be buying in Google.” (26:52, Matt Britton)
• Content Generation at Scale: Transcripts are anonymized, summarized, and repurposed into blog posts (with sensitive info redacted), scheduled and published automatically. These posts drive both organic SEO and feed into dynamic search ads.
  • “We have 10,000 blog posts that are created on the calls that we're making without any human intervention.” (35:25, Matt Britton)
  • “It'll take those transcripts and it'll write a blog post that fully redacts all that specified information, but focuses just on the idea.” (35:25, Matt Britton)
• Database Enrichment: Structured customer profile database is automatically populated from transcript data, providing sales with actionable insights before future calls.

4. Data Infrastructure

• Feeds structured information into both Google Sheets and Salesforce (with recognition that custom structures are often preferable to rigid schemas).

Notable Quotes & Memorable Moments

The Power of “Doing It Yourself”

Matt Britton (09:35):
“You'll go nowhere. Yep. No, you'll go nowhere. And I've said this a lot. This is a moment for actual hard skill building in leaders, which is you actually have accessible skills to build in using AI...to really upskill yourself on the capability. And that's going to make you a much more relevant leader.”

Building Value from the “Wild”

Matt Britton (33:58):
“People are so focused on the application layer, it means nothing without the data. And to me it's like this is the ultimate source of data and this is the treasure trove and this is people in the wild saying what they want. So I want to build everything on top of this data.”

Automating at Scale, Not in Silos

Ariel (34:43):
“Think of yourself as a single workflow, think of your team as a single workflow, maybe even think of your company as a single workflow and figure out how that whole thing should work...it's really interesting that you have this mega automation as opposed to these little one off things.”

Democratizing Coaching & Reducing Human Error

Matt Britton (30:25):
"AI is an incredible tool because [the good reps are] going to want this feedback. And the people who never really wanted to hear from anyone to begin with, they're not going to want to hear this. But they wouldn't have been good in either way. So that kind of goes to the point that like it's going to make the good people that much better."

Organizational Impact & Future of Teams

Matt Britton (39:45):
“I think it's far more individual contributors, far more people who want to put their hands on keyboard, people who are willing to learn...Not order takers, not people who walk into work every day and wait to be told what to do. Because...I don't need more people to tell what to do. I need people who are going to come up with new ideas and solutions and be proactive.”

Timestamps for Important Segments

| Timestamp | Topic | |-------------|-------------------------------------------------------------------| | 00:00–04:32 | Matt describes the root problem and discovery of call recordings | | 09:02–14:01 | Building and hacking the workflow with Zapier and Browse AI | | 17:16–23:25 | Summarization prompt, LLM choice, and orchestration | | 23:25–26:17 | Slack summaries, sentiment scoring, and churn alerts | | 26:52–28:34 | Keyword extraction for paid search | | 28:34–30:25 | Automated coaching, feedback systems, and follow-up emails | | 32:28–33:58 | Data infrastructure: CRM vs. custom structured databases | | 34:43–35:25 | Aggregation, "single workflow" philosophy | | 35:25–37:51 | Scaled content generation and marketing automation | | 39:45–41:06 | Impact on team structure and roles | | 41:21–41:49 | Matt’s Advanced Prompting Framework |

Takeaways for Listeners

• Focus on the Data: Leverage your organization’s “real voice of the customer” as the foundational source for automation.
• Get Hands-On: Leaders should learn no-code AI tools to unlock new operational possibilities and ask better of their technical teams.
• Automate for Scale: Tightly integrated automations can transform a single data asset into outputs for sales, marketing, and ops in real time.
• Democratize Insights & Coaching: Automated feedback, early warning systems, and customer summaries make organizations more transparent and responsive.
• New Roles Emerging: The future team needs “go-to-market orchestrators” plus proactive ICs—automation generalists and specialists, not just order takers.

Where to Learn More

• Matt Britton: mattbritton.com | Book: Generation AI (May)
• Suzy: suzy.com
• Podcast episodes: howiaipod.com

(Summary created by parsing, synthesizing, and structuring the core content of the How I AI podcast episode, maintaining the direct tone and insights of the conversation. Ads and non-content sections omitted.)