Voice of Customer Research: How PMMs Build Messaging That Actually Converts

The PMO director had seen the demo three times.

She'd sat through a 45-minute walkthrough, a QBR presentation, and a personalized trial. The product was genuinely good — the AE knew it, the SE knew it, and she seemed to know it too.

And then she went dark.

Three weeks later, on a win/loss call, she finally explained what happened. "The product was fine," she said. "But every time your team talked about it, they talked about the platform. The integrations. The configurability. What I needed to hear was whether you understood what was happening in my org. My VP is asking me to do 40% more with the same headcount. I needed to know if your product understood that problem."

The AE had given her a perfect product demo. He'd never given her a single line that said: we understand why you're under pressure.

This is the gap that Voice of Customer research closes.

---

Most B2B messaging is written from the inside out. It starts with the product — what it does, how it's built, which features differentiate it from competitors — and works outward toward some assumed buyer. The result is marketing copy that sounds like it was written by the people who built the thing, because it was.

Buyers aren't confused by this messaging. They're just unmoved.

The alternative is outside-in messaging. Messaging that starts with the buyer — their real problems, their actual language, the emotional context of their decisions — and works backward toward the product. When you do this well, buyers feel like your product was built specifically for them. Not because it was. Because the way you talk about it proves that you understand what they're dealing with.

Voice of Customer research is how you get there.

What VoC Research Actually Is (and Isn't)

Voice of Customer research is the systematic practice of capturing how buyers describe their own problems, motivations, and purchasing decisions — in their own words.

It's not a survey. It's not an NPS score. It's not a focus group where you show concepts and collect reactions.

VoC is conversation, followed by pattern recognition. You talk to buyers — wins, losses, churned customers, near-misses — and you listen for the specific language they use to describe what they were dealing with, what they were hoping for, and how they made the decision they made. Then you systematize what you heard and use it directly in your messaging.

The output isn't a slide deck. It's a language bank — a structured library of customer quotes organized by theme, ready to pull from when you're writing website copy, battle cards, email sequences, or pitch decks.

Done well, VoC research answers three questions your product roadmap can't:

1. What are buyers actually trying to accomplish? (Not "what does your product do" — why does anyone care?)
2. How do they describe the problem before they know a solution like yours exists?
3. What language, specifically, makes them feel understood?

The PMM's VoC Toolkit

There are five primary sources of VoC data. You don't need all of them to start — but you need more than one.

1. Win/Loss Interviews

Win/loss interviews are the highest-signal VoC source available to a PMM. If you do nothing else, do this.

The structure matters. A win/loss interview is not a customer success check-in or a post-sale debrief. It's a structured conversation designed to understand the decision — specifically the moments that shaped it.

For a win interview, the core questions are:

• Walk me through what was happening in your business right before you started looking for a solution. What changed?
• How did you define what you were looking for? What were your evaluation criteria?
• Who else were you evaluating? What was it about them that made them alternatives?
• Was there a moment when you felt like we understood your problem? What happened?
• What almost made you go a different direction?

For a loss interview (the hardest and most valuable), the questions are:

• Where did we lose you?
• What did the winner say or do that we didn't?
• Is there anything we could have shown you that would have changed the outcome?
• Looking back, was there a moment where you started to pull away?

Most PMMs never ask these questions because win/loss interviews require a process — someone has to recruit participants, schedule the call, and ensure the AE doesn't poison the well by having a defensive conversation first. Build the process. It's worth it.

A useful heuristic: aim for 5 win interviews and 5 loss interviews per quarter per major segment. At that volume, you'll see messaging patterns emerge within two to three cycles.

2. Customer Advisory Boards and Executive Interviews

Your most engaged customers are walking repositories of the language that sells your product. They've been through the buying process. They remember why they chose you. They understand the alternatives. And if you've built the relationship correctly, they'll tell you exactly what they were thinking — including the parts the sales team never heard.

CABs are a structured way to access this at scale. Executive interviews (one-on-one, 30 minutes, once or twice a year) are more intimate and often more honest.

The questions to ask that most PMMs skip:

• How do you describe what you use us for when you're explaining it to someone who's never heard of us?
• What was the business case you made internally to get approval?
• If you were evaluating us again today, what would you want to see first?

That second question — the internal business case — is gold. It tells you how your customers are already selling you inside their organizations. If you're not using that language on your website, you're making their job harder.

3. Support Tickets and Product Reviews

Every support ticket is a sentence about what a customer expected vs. what they got. Every G2 or Capterra review is a mini win/loss interview written in public.

Mining these sources won't give you the depth of a live interview. But it gives you volume and unfiltered language — nobody's being polite in a support ticket.

What to look for in support tickets:

• Frustration language: the specific words customers use when something doesn't work as expected. This is where you find objections before they happen.
• Workaround descriptions: "I've been doing X manually because I can't figure out how to Y." These reveal job-to-be-done gaps your product hasn't addressed.
• Expectation statements: "I thought this was going to..." — these tell you where your marketing promised something your product didn't deliver.

What to look for in public reviews (G2, Capterra, App Store, Trustpilot):

• The headline a 5-star reviewer writes: often the most distilled articulation of your value proposition, in customer language.
• The specific pain the reviewer mentions being solved: this is the JTBD trigger moment, described retrospectively.
• The specific complaint in a 3-star review: these are objections your sales team is handling (or failing to handle) live.

Build a process for this. Have someone read through 50 recent reviews and tag every phrase that describes a pain point, an outcome, or a frustration. You'll find more messaging gold than in most internal brainstorms.

4. Jobs-to-be-Done Interviews

Jobs-to-be-Done is a framework developed by Clayton Christensen that reframes the question from "what does the customer want" to "what are they trying to accomplish." The famous example: people don't buy a quarter-inch drill. They buy a quarter-inch hole. More precisely: they buy a way to hang a picture so their family feels at home.

JTBD interviews are structured differently from win/loss interviews. The goal isn't to understand the purchasing decision — it's to understand the triggering event. What happened right before the buyer started looking?

The timeline interview is the most effective format:

• Pre-event: "Take me back to before you started looking for a solution. What was going on in your work? What was the situation?"
• The trigger: "Was there a moment when you decided you needed to do something? What happened? What did that feel like?"
• The search: "What did you search for first? What were you calling the problem at that point?"
• The alternatives considered: "What else did you try or consider? What made you put it aside?"
• The switch: "What made you pull the trigger on this?"

The insight JTBD interviews consistently generate: your buyers often don't use your category language when they start searching for a solution. A PMM at a project management company once discovered that most of her buyers weren't searching for "project management software" — they were searching for "how to stop my team from missing deadlines." That's a headline. That's homepage copy. That's the language that makes someone feel seen.

5. Sales Call Review

Your sales team is running VoC research every single day. They just aren't calling it that, and they aren't systematizing it.

Gong, Chorus, and even basic call recordings give PMMs access to the unfiltered buyer voice in a live sales context. The objections are real. The competitor mentions are unscripted. The language buyers use to describe their problem is exactly what they'd type into a search bar or say to a colleague.

Make it a practice to listen to 4 to 6 sales calls a month — not to evaluate the AE, but to mine for buyer language. Specifically:

• How does the buyer describe their current situation in their own words?
• What objection comes up most often in discovery?
• What question does the buyer ask that the AE seems unprepared for?
• When does the buyer's tone change — and what caused it?

This is free VoC research your company is already generating. Most PMMs don't tap it because it requires building a relationship with the sales team and carving out time. Both are worth it.

Building a Language Bank

The output of VoC research isn't a summary slide. It's a language bank — a structured repository of customer quotes, organized so you can pull from it when you write.

A simple language bank has four columns:

The goal is specificity. Not "customers care about ease of use" — that's a conclusion. The language bank captures the exact sentence that surfaced the insight: "I'm not going to pay for a tool my team won't use because it takes an hour to learn."

When you have 50 to 100 quotes organized this way, you have something valuable: a messaging brief written by your customers. Every homepage headline, every email subject line, every sales deck title is a potential pull from this bank.

From Language Bank to Messaging

Here's where most VoC efforts stall. The research gets done. The quotes get collected. And then nothing happens, because the translation from insight to copy isn't obvious.

Here's a practical approach.

Step 1: Identify the top three pain themes.

Look at your language bank and count which themes appear most often. If 11 out of 20 interview quotes reference the same underlying frustration, that frustration is your headline problem. It goes at the top of the page, in the opening paragraph of every piece of content, and in the first sentence of your email sequence.

Step 2: Find the trigger language.

Your buyers started looking for a solution when something happened. Identify the triggering event language from your JTBD interviews: "when the team grew to 20 people," "after we missed a major launch deadline," "when we started getting questions from the board about metrics we couldn't answer." This language belongs in your demand gen — you're meeting buyers at the moment they recognize they have the problem.

Step 3: Steal the outcome language.

How do your happiest customers describe what life looks like after using your product? This is your value proposition in their words. Not "real-time visibility into project status" — whatever they actually said when you asked them what changed. "I stopped getting the 10 PM Slack messages asking where things stood." Build your hero section around the outcome, not the feature.

Step 4: Run the outside-in test.

Take a piece of copy you've already written and highlight every phrase that came directly from a customer interview, support ticket, or review. If less than 40% of it is highlighted, it's inside-out. Rewrite the unhighlighted sections using language from your bank.

Step 5: Test and update.

The best messaging test is whether buyers engage with the language differently. Watch email open rates, ad CTRs, and sales call transcripts for signs that the language is resonating — buyers repeating it back, spending longer on specific pages, asking questions that suggest they already understand your positioning. VoC is a loop, not a one-time project. The best PMMs refresh their language bank quarterly.

Common VoC Mistakes (and How to Avoid Them)

Confusing satisfaction with insight. A customer who says "everything is great" and gives you a 9/10 is not a VoC interview. The useful interviews are the ones where you push past satisfaction into specifics: what specifically is great? What would make it a 10? What would need to go wrong for you to leave?

Only interviewing happy customers. Your happiest customers confirm your assumptions. Your churned customers and lost deals challenge them. The insights that move messaging tend to come from the people who didn't stay.

Paraphrasing instead of quoting. The point of VoC research is the language. When you paraphrase a customer quote into clean marketing language, you've already filtered out the authenticity. Keep the raw quote. Use it.

Treating it as a one-time project. Buyer language evolves. Your market moves. The language that resonated in Q1 2024 may be flattened by Q4 2025. VoC is a practice, not a project. Build it into your quarterly rhythm.

Not closing the loop with sales. The output of VoC research should feed directly into your sales enablement program. The objections you hear in loss interviews become battle card content. The trigger language you capture in JTBD interviews becomes discovery questions. VoC without sales activation is research in a drawer.

The Bigger Picture

There's a reason that the best product marketers in B2B SaaS treat customer research as a core PMM skill — not a nice-to-have.

Positioning, messaging, and competitive differentiation are all downstream of one question: do you understand your buyers better than your competitors do? If you do, your messaging will sound like it was written by someone who's lived the problem. If you don't, it will sound like it was written by someone who built the solution.

Buyers can tell the difference. They may not be able to articulate why one company's copy resonates and another's doesn't. But they feel it. "These people get it" is the feeling that drives trust, and trust is what drives the deal.

Voice of Customer research is how you earn that feeling. Not by being clever. By listening.

---

Bare Strategy helps B2B SaaS companies build product marketing foundations that last — positioning, messaging, and go-to-market strategy grounded in research, not assumptions. If your messaging needs a reset, let's talk.

---

Related reading:

• Win/Loss Analysis: The PMM's Most Underused Revenue Weapon
• From Features to Value: How to Write Messaging That Moves Deals
• Why Most Positioning Fails (And the Framework to Fix It)
• Sales Enablement That Actually Enables Sales