Using Voice Data for Market Analysis

How PE firms should be using AI (Edition 001) Let’s say you have a home services company and want to understand pricing across every market. Traditional methods are manual and inconsistent. Someone calls around for quotes, maybe hits 5-10 if they're diligent and take notes in a spreadsheet. The data is spotty, quickly outdated, and rarely comprehensive enough to inform real strategy. With AI, voice agents can call every HVAC, plumbing, and electrical competitor within a 10-mile radius of each service area, request quotes for standard jobs (AC tune-ups, water heater installs, 200-amp panel upgrades), capture pricing details along with booking availability and service timeframes, and map them against the company's own rates. The agent handles the full conversation, explains the job requirements, asks follow-up questions, and even navigates gatekeepers or callback requests. In less than a day, you can determine exactly which markets you're under- or over-priced in, identify if certain service categories are consistently misaligned with competition, and see if pricing gaps correlate with win rate differences or conversion metrics between regions. You might discover that your water heater pricing is competitive in Phoenix but 20% above market in Dallas, or that competitors in certain zip codes are offering same-day service at lower rates... insights that directly explain why some territories are underperforming. This is the kind of competitive intelligence that would typically require a dedicated ops person spending weeks on the phone, or the kind of market analysis a consulting firm would bill for over a multi-month engagement. Instead, it's actionable insight delivered at scale, refreshable on demand, and granular enough to inform market-level pricing strategy.

B2B has a brand measurement advantage over B2C marketers. We just haven’t been using it to its full potential, and I don’t know why. Here's one things I've loved learning over the past 8 years in my role. I can hear on calls when someone has been familiar with our brand, I can hear why they reached out, and I can listen for tells that indicate what other agencies they've talked to. B2B should be leveraging call data more to learn about their brand and how they are being perceived. In consumer markets, you need surveys, focus groups, and social listening to get a read on brand perception. In B2B, we get something far more valuable: direct conversations with prospects and customers, every single day. Think about it: 📞 Account manager catch-ups reveal how customers are discovering and adopting new features. 📞 Sales calls uncover how prospects first heard about you, what sparked their interest, and who they’re comparing you to. This is brand intelligence that comes straight from the source. And now, with AI, we can scale it: 🎧 Transcribe and analyze hundreds (or thousands) of calls. 🎧 Detect sentiment shifts over time. 🎧 Spot the words, competitors, and triggers that signal brand strength (or weakness). Real examples of what’s possible: ☎️ Listen to call transcriptions to see if prospects and customers are talking about your branded products and terms. ☎️ Understand if they’re asking for things unique to you—or unique to your competitors. ☎️ Recently launched a brand initiative? See if they’re using the language your campaign focused on. We can analyze these calls to get brand sentiment, track brand lift, and capture feedback... without running a single extra survey. This is one area in research and insights where GTM engineering can compress product marketing cycles and turn what used to be qualitative data into quantitative.

Most marketers are missing the biggest goldmine of attribution data: What customers actually say on calls. Yes, it’s a lot of unstructured data sitting in a place that marketers can’t access. But that’s the perfect use for AI. I sat down with Justin Norris this week on Stack & Scale, and he showed me exactly how he fixes this with an AI workflow. He is turning Gong calls into a real-time attribution and insights engine. Here’s how: – Pull call transcripts from Gong – Run them through AI to extract competitor mentions, objections, and product feedback – Pipe the structured data into Snowflake for analysis – Match it back to campaigns and revenue Now, instead of guessing why deals move, Justin’s team can tie marketing influence directly to customer conversations. No, it’s not perfect (no attribution solution is), but it’s closer to the truth than any attribution software on the market. In our conversation, we also covered: – Why reliability is the hardest part of AI in production – The stack Justin uses to scale this (Retool, GPTs, Gemini, Claude, Dust) – When you should move beyond Zapier for serious automation Listen to the full episode with Justin Norris and get his entire workflow to copy + paste on my Stack & Scale Substack.

Took me a few months to notice. But my best Zapier workflow was missing something BIG. I built the Voice of Customer workflow and database earlier this year to pull EVERYTHING I could from sales calls. Pain points, customer quotes, buying triggers, tooling, etc. etc. All the stuff that we as marketers KNOW is gold but normally just sits in transcripts your team might not even have access to. I thought it was PERFECT. I was wrong. Here's what I left out: What prospects are saying about your COMPETITORS. Sure, I grabbed what competitors were mentioned in each call. It made a nice list to look at, but that was about it. I was so obsessed with extracting customer pain points, quotes, and questions, that I completely ignored a key part of the convo. The part where prospects are literally telling you why they're leaving their current tool. What amazing feature broke on them for the 10th time. Why the pricing didn't make sense as they grew. How a new VP is looking to shake things up. All the good stuff. If you attended last week's webinar with Fellow - AI Meeting Assistant, you got a sneak peak of the Anti-VoC Database. If not, here's the 411. The Anti-VoC database is organized by COMPETITOR instead of by call, which honestly makes it so much more useful. My main VoC Databases are broken down by prospect or company. When you need to know what prospects said about a specific competitor, you're digging through those rows. So to flip that on it's head, the Anti-VOC Database creates individual rows for every competitor mentioned in a call. So if a call mentions 4 competitors, there are 4 new rows created. With all the context you could ever need about how a prospect views your competitors and the market in general. Plus, some general notes on the prospect and company. ALL IN ONE SPOT! After a week or so of this running in the background, you will have a VERY strong competitive intel database. Ready to inform your next landing page build or spark a brand new ad campaign. Your prospects are telling you exactly how to stand out in the market. With the Anti-VoC Database, you can start to listen. TBH This might be my new favorite build, and I hate that it took me this long to share it with the world. 👋 Check out the workflow preview in this deck, and DM me for the entire build guide if you want to see how to build it from scratch.

I see nobody doing this with AI voice agents. So I did. This is unlocking a whole new layer of intelligence on your AI voice calls. What is it? Sentiment Anlalysis. Why does this matter? Because most businesses are sitting on a goldmine of voice data... but they’re not extracting the emotional signals that drive real outcomes. Here’s where sentiment analysis actually adds value: ✅ Customer Experience Monitoring Spot unhappy customers early. Trigger an automatic follow-up if a call turns negative. ✅ Agent Performance Tracking See how sentiment shifts across reps, scripts, or time. Is your team actually creating positive experiences? ✅ Trend Recognition Negative sentiment = higher churn? Now you've got predictive insights. ✅ Training & QA Flag poor sentiment calls for review. Let AI highlight the moments that caused friction. But it's not always worth your time... Sentiment analysis is useless if: → You're not acting on the data. → Your calls are too short or robotic. → You don’t have enough volume to find patterns. → Your domain needs custom sentiment tuning (sarcasm, mixed languages, etc.). Want to make it actually useful? Here’s how: → Link sentiment to outcomes like conversions or renewals. → Create real-time alerts or dashboards for your team. → Fine-tune the model on your transcripts, not generic ones. → Combine it with other signals like talk-time, interruptions, and keywords. The emotional layer of your calls is where the real insight lives. Curious how this works in practice? I’m happy to show what I built today. Drop a “curious” below or shoot me a message.

Voice is the new Big Data! On our Momentum Atlanta Microsoft Roadshow Call with Brandon B., James Fox and Elizabeth Backus Hildreth, MBA talking about how structured customer data is captured in Microsoft Teams. For decades, we've treated business phone calls like a utility. A line item on a budget. In the next 24 months, that thinking will become obsolete. We meticulously analyze clicks, opens, and web traffic. Yet we've completely ignored the most valuable data source in the entire company. Raw, unfiltered customer conversations. 🗣️ Every single day, a goldmine of business intelligence is created on your calls. Customer objections. Competitor mentions. Buying signals. And then it disappears the second your agent hangs up. But that’s about to change. With AI like Microsoft Copilot layered on a modern Teams Phone system, we can finally analyze every conversation at scale. 🤖 Voice is no longer just a conversation. It's structured data. 📊 The future isn't just unifying your communications; it's about making them intelligent. Your phone system is about to become your most powerful BI tool. 💡 Imagine: 1. Real-time sales coaching triggered by keywords in a live call. 2. Automatically identifying at-risk customers from sentiment analysis. 3. Informing your entire product strategy with every feature request. The next 24 months will see a dramatic shift from AI as a feature to AI as the core fabric of the Teams OS. Voice data, transformed by AI, will become a primary source of business intelligence. Stop managing your phone system as a cost center and start leveraging it as your next big data platform. If you could automatically analyze 100% of your customer calls, what is the #1 question you would want to answer?