How I AIHow this CEO turned 25,000 hours of sales calls into a self-learning go-to-market engine
Claire Vo and Matt Britton on cEO turns Gong call transcripts into automated GTM intelligence engine.
In this episode of How I AI, featuring Matt Britton and Claire Vo, How this CEO turned 25,000 hours of sales calls into a self-learning go-to-market engine explores cEO turns Gong call transcripts into automated GTM intelligence engine Suzy CEO Matt Britton describes a “mega workflow” that starts with a newly completed Gong-recorded call and automatically scrapes the transcript, cleans it, enriches it with internal data, and runs multiple LLM analyses.
At a glance
WHAT IT’S REALLY ABOUT
CEO turns Gong call transcripts into automated GTM intelligence engine
- Suzy CEO Matt Britton describes a “mega workflow” that starts with a newly completed Gong-recorded call and automatically scrapes the transcript, cleans it, enriches it with internal data, and runs multiple LLM analyses.
- The workflow posts structured call summaries and sentiment scores to Slack, flags churn risk in an early-warning channel, and generates coaching feedback plus a human-in-the-loop follow-up email draft for the rep.
- On the marketing side, the same call data produces Google Ads keyword suggestions, builds an aggregate customer-profile database for retrieval, and can generate fully redacted SEO blog posts published later to protect confidentiality.
- A recurring theme is leadership being hands-on: Britton argues executives should learn to build with no-code/AI tools, focus first on the core business problem and data source, and hire “super ICs” who proactively orchestrate and improve automations.
IDEAS WORTH REMEMBERING
5 ideasStart with one painful business problem, not a list of AI tools.
Britton’s trigger was sales/CS saying they “couldn’t find anything.” He recommends stepping back to identify what’s blocking growth, then building AI around that specific constraint.
Your best AI engine is often data you already have—especially customer conversations.
Suzy had ~25,000 hours of recorded calls. Britton treats transcripts as the highest-fidelity “people in the wild” dataset for product, sales messaging, churn risk, and marketing demand.
When native integrations don’t expose what you need, scraping and “hacks” can unlock the workflow.
Gong didn’t provide an easy transcript feed, so he used the call ID pattern + Browse.ai to scrape the transcript URL and push it into Zapier as the automation trigger.
Operational reliability matters: small steps like delays and HTML stripping prevent automation breakage.
He adds a 1–2 minute delay to ensure the scrape completes and uses formatting to remove HTML/noise before sending transcripts into LLM prompts.
LLM summaries become more valuable when they’re structured and scored.
The workflow produces a standardized call overview plus a 1–10 sentiment score; Suzy benchmarks sentiment trends against real churn/upsell outcomes to make it predictive, not just descriptive.
WORDS WORTH SAVING
5 quotesI always knew we had Gong, but what I didn't know is that their transcripts were amazing, and that we actually had 25,000 hours of call transcripts… there's no better source of truth.
— Matt Britton
It wasn't connected. I had to kind of hack it together.
— Matt Britton
It is not sufficient to instruct your engineers to build AI… you'll go nowhere.
— Claire Vo
It’s not about the tool, it’s about the data… this is people in the wild saying what they want.
— Matt Britton
I need people who are gonna come up with new ideas and solutions and be proactive.
— Matt Britton
QUESTIONS ANSWERED IN THIS EPISODE
5 questionsOn the scraping step: what specific Gong limitations forced you to use Browse.ai, and how do you monitor/handle breakage when Gong’s UI changes?
Suzy CEO Matt Britton describes a “mega workflow” that starts with a newly completed Gong-recorded call and automatically scrapes the transcript, cleans it, enriches it with internal data, and runs multiple LLM analyses.
How did you validate that the 1–10 sentiment score is “highly predictive” of churn—what was the methodology and what thresholds/false positives did you see?
The workflow posts structured call summaries and sentiment scores to Slack, flags churn risk in an early-warning channel, and generates coaching feedback plus a human-in-the-loop follow-up email draft for the rep.
What does the churn early-warning routing logic look like (e.g., below 7), and how do you separate “unhappy with their business” from “unhappy with Suzy”?
On the marketing side, the same call data produces Google Ads keyword suggestions, builds an aggregate customer-profile database for retrieval, and can generate fully redacted SEO blog posts published later to protect confidentiality.
For the coaching feedback: what signals does the model use (interruptions, talk time, missed topics), and how do you prevent unfair or misleading evaluations?
A recurring theme is leadership being hands-on: Britton argues executives should learn to build with no-code/AI tools, focus first on the core business problem and data source, and hire “super ICs” who proactively orchestrate and improve automations.
What guardrails do you use to ensure PII and confidential strategy details are redacted before auto-publishing blog content, and what was your testing process before going live?
Chapter Breakdown
Sales team can’t find answers—25,000 hours of calls become the source of truth
Matt Britton explains the core pain: sales and CS teams lacked easy access to customer insights and patterns. The breakthrough was realizing the company had amassed ~25,000 hours of recorded Gong calls, creating a high-signal dataset for a self-learning go-to-market engine.
Why Zapier became the automation backbone (and the AI tipping point)
Matt shares why Zapier works for him as a non-coder who likes stitching tools together. Once Zapier integrated AI/LLMs, it expanded from simple glue automation into an orchestration layer for sophisticated GTM workflows.
Start with the business problem, then choose tools and data
They emphasize avoiding “tool wandering” and instead identifying the single constraint holding growth back. Once the problem is clear, the right data sources and automations become obvious—calls were the missing link.
CEO hands-on building: the leadership skill shift in the AI era
Matt and Claire argue leaders must build and understand workflows themselves, not just delegate to engineering. Hands-on automation teaches practical AI fluency, improves judgment, and reduces dependency on “black box” estimates.
Triggering on a new Gong call: hacking the call ID feed with Browse AI
Matt walks through the hardest early step: getting a reliable trigger that captures each new call and its unique ID. Because Gong didn’t expose an easy path, he used URL patterns plus Browse.ai scraping to pull transcripts programmatically.
Cleaning and enriching transcripts: delays, formatting, and lookups
After scrape completion, the automation buffers with a short delay to prevent errors, then cleans HTML and normalizes text. Matt enriches the record by pulling additional context from Google Sheets and other systems to “round out” missing data.
LLM selection and operations: model choice, stability, and cost tradeoffs
Matt explains how he chooses models pragmatically: don’t change what works, but test outputs in ChatGPT across models for speed and quality. He also flags the operational challenge of maintaining many automations without perfect handoffs and documentation.
Core Summary Generator: call overview, sentiment scoring, and next steps
A central LLM prompt transforms each transcript into an actionable brief: participants, objectives, outcomes, sentiment, what went well, improvement areas, and next steps. The sentiment score becomes a measurable signal that can be benchmarked against churn and expansion outcomes.
Real-time Slack visibility: company-wide call intelligence + churn early warnings
The workflow posts summaries into Slack to create a live feed of customer reality—useful even for a 300-person company. Low sentiment scores route into a churn early-warning channel so leadership can intervene before problems become churn events.
From customer language to demand gen: extracting keywords for Google Ads
Matt shows how each call also fuels marketing: an LLM extracts high-intent terms customers actually use. Those terms are automatically added to Google campaigns, tightening the loop between voice-of-customer and paid acquisition.
AI coaching for Sales/CS: individualized feedback and trend tracking
The automation generates immediate coaching notes for the rep: strengths, weaknesses, and improvement suggestions. This feedback is also stored so managers can spot behavior patterns and make performance reviews more objective and continuous.
Human-in-the-loop follow-up email writer: faster, better post-call execution
To reduce the burden of post-call admin, the system drafts a high-quality follow-up email the rep can copy, edit, and send. Matt keeps a human approval step to prevent misfires and allow contextual judgment (timing, recipients, tone).
Building an aggregate customer profile database for RAG and prep
Beyond per-call actions, the workflow structures each call into a database: roles, product interests, trends, and use cases. Sales can then query aggregated patterns (e.g., what automotive brand managers care about) to prep smarter and standardize playbooks.
Redacted content engine: auto-generating SEO blog posts and ads from calls
The most ambitious workflow converts calls into anonymized, SEO-optimized blog posts that remove all identifying details. Posts publish after a delay (e.g., 21 days), scale into thousands of pages, and can feed dynamic search ads—turning conversations into reusable market-facing assets.
Org design implications and prompting approach: super ICs + guardrails
Matt closes on how automation changes hiring and ownership: fewer “order takers,” more proactive builders and operators, plus a small group of GTM automation orchestrators. For prompting, he uses a guardrail-first method—clearly stating goals and what the model must not do, then iterating toward the desired output.
EVERY SPOKEN WORD
Install uListen for AI-powered chat & search across the full episode — Get Full Transcript
Get more out of YouTube videos.
High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.
Add to Chrome