PROMPT This AI Challenge

Episode 1 - AI, Strategy & the Unfiltered Truth with Charles Hicks

Every episode of the PROMPT This podcast includes an AI Challenge for the audience.  Follow the instructions below to complete this episode's challenge.

If you’re sitting on a pile of Zoom transcripts from prospect calls, you’re not alone. Every company with a halfway decent sales motion has hours of recordings collecting digital dust. The issue? Most organizations don’t squeeze the juice out of them. Those transcripts are a goldmine for uncovering prospect pain points, objections, competitor mentions, and the subtle buying signals that don’t always make it into your CRM.

AI can help turn that pile of text into actionable sales intelligence. This episode's AI Challenge is to take your existing customer or prospect call transcripts and pass them through your favorite LLM like ChatGPT, Google Gemini or MS CoPilot.  We used ChatGPT for our practice session since this is what our guest, Charles Hicks, recommended.

We break the challenge down into a few different approaches: first, the old-fashioned manual way, then the scalable automated way, and finally, using SaaS platforms purpose-built to do this at scale.

Approach #1: The ChatGPT Manual Playbook

Before automation, you need to understand the anatomy of the process. Here’s how you’d do it by hand, with ChatGPT as your assistant:

1. Collect Your Transcripts
   Export Zoom call transcripts into text or Word files. Organize them by date, account, and rep.
   
   
2. Chunk the Data
   Extremely long transcripts overwhelm AI models. This means you can’t paste every transcript you have into one prompt.  ChatGPT can theoretically handle up to 15 hours of call transcripts in one prompt or request/response cycle.  Paste the conversation transcripts into ChatGPT in manageable chunks (10–15 conversation transcripts per paste depending on length).
   
   
3. Define What You’re Looking For
   Give ChatGPT specific instructions, such as:
   
   
• List all pain points mentioned by the prospect.
  
  
• Extract objections to pricing, timing, or implementation.
  
  
• Identify competitor mentions.
  
  
• Summarize the prospect’s buying criteria.
  
  
4. Aggregate Across Calls
   After analyzing several transcripts individually, compile the outputs into a spreadsheet. This lets you spot patterns like “pricing objections came up in 60% of calls” or “Competitor X is mentioned in 4 out of 10.”
   
   
5. Synthesize Into Insights
   Summarize the themes across transcripts:
   
   
• Top three pain points prospects care about.
  
  
• The objections that consistently stall deals.
  
  
• Competitive intelligence to arm your reps.
  
  

This manual process works if you’ve got forty to fifty transcripts. But if you’ve got two hundred fifty? You’ll drown in copy-paste. That’s where automation steps in.

Approach #2: Automating the Analysis with ChatGPT APIs

Automation is where ChatGPT becomes less a helpful assistant and more like a junior analyst working 24/7. Here’s the playbook:

1. Use a Transcript Repository
   Store all Zoom transcripts in a shared folder or database. Google Drive, OneDrive, or Dropbox work fine. More advanced teams push transcripts into a CRM or data warehouse.
   
   
2. Set Up Preprocessing
   Scripts can split transcripts into clean, structured chunks (e.g., per speaker, per 2000 words). This avoids pasting walls of text into ChatGPT.
   
   
3. Create Prompt Templates
   Build reusable prompts that direct ChatGPT to extract:
   
   
• Pain points
  
  
• Objections
  
  
• Competitor mentions
  
  
• Buying signals
  
  
4. Leverage Workflow Tools
   
   
• Zapier / Make.com: Trigger analysis every time a new transcript lands in your folder.
  
  
• LangChain or LlamaIndex: Run structured queries across all transcripts at once.
  
  
• Google Sheets / Airtable: Push extracted insights into structured tables so sales leaders can track trends.
  
  
5. Generate Dashboards
   With all data centralized, you can build dashboards that show:
   
   
• Frequency of each pain point by industry.
  
  
• Objections by deal stage.
  
  
• Competitor mentions over time.
  
  

Approach #3: SaaS Platforms That Do This For You

If you’d rather skip the build and go straight to off-the-shelf call analysis, several SaaS platforms specialize in analyzing prospect calls and spitting out insights. Here are the heavy hitters:

• Zoom Revenue Accelerator
  Zoom has built its own AI-driven conversation intelligence into the platform. It automatically generates meeting summaries, captures action items, analyzes sentiment, and flags potential risks in deals. Because it’s native to Zoom, there’s no need for extra integrations—making it especially appealing to teams already running their calls through Zoom every day.
  
  
• Gong
  The category leader. Gong records, transcribes, and analyzes calls to surface deal risks, competitor mentions, and rep talk-time ratios. Its “deal warnings” feature is unique—flagging opportunities at risk before they die.
  
  
• Chorus (now part of ZoomInfo)
  Similar to Gong but integrated tightly into ZoomInfo’s go-to-market stack. Chorus excels in coaching reps, highlighting top objections, and enabling snippet sharing for training.
  
  
• Wingman (now Clari Copilot)
  Now part of Clari. Strong real-time coaching tool—pops up suggestions mid-call when a rep faces objections. Its strength is less post-call reporting and more “in the moment” enablement.
  
  
• Avoma
  A versatile player. Avoma covers meeting notes, summaries, and even agenda preparation. It’s favored by SMBs for ease of use and affordability compared to Gong/Chorus.
  
  
• Salesloft Conversations
  Integrated into Salesloft’s broader engagement platform. Great for aligning call insights directly with cadences, making it a strong pick for outbound-heavy teams.
  
  
• ExecVision (now part of Mediafly)
  Now part of Mediafly.  Known for deep coaching analytics. It goes beyond “what was said” to focus on rep behavior patterns over time, helping sales leaders design coaching programs.
  
  
• Fireflies.ai
  More of a lightweight, affordable solution. Great for startups that just want reliable transcription, searchable conversations, and basic AI summaries.
  
  

Each tool has its niche. The easy answer is Zoom Revenue Accelerator if you’re using Zoom already.  Enterprise teams with deep pockets lean toward Gong or Chorus. Mid-market players often like Avoma or Wingman. Startups dip their toes with Fireflies.

AI Prompts

Here are examples of prompts you can drop into ChatGPT to run manual or automated transcript analysis.  Be sure to upload your call transcripts into the prompt window so that ChatGPT knows to analyze those files. Or create a ChatGPT project and upload the files only once.

• Analyze this transcript and list all prospect pain points mentioned.
  
  
• From this transcript, summarize every objection raised and categorize them into pricing, timing, or implementation.
  
  
• Extract any competitor names mentioned in this conversation and note the context.
  
  
• Summarize the top 3 buying criteria the prospect seems to value.
  
  
• Compare these transcripts and highlight common pain points and objections across all.
  
  

Final Thought

Sales calls are one of the most underutilized datasets in your company. Manually, ChatGPT can help you dig insights from a handful of calls. Automation takes it further, turning every conversation into structured intelligence that sharpens your sales pitch, arms your reps, and feeds your product roadmap. And if you don’t want to roll your own solution, SaaS tools like Gong, Chorus, and Avoma can do the heavy lifting for you.

The difference between companies who “record calls” and those who “act on calls” is the difference between guessing and winning.