Voice of Customer Scan

• Source: Built by Jonathan Kvarfordt on the GTM AI OS. Supplied as a complete zip and served unchanged.
• Frameworks applied: Role scans

It covers a job one of the eight revenue seats has to finish, and it states its process, decision rules, and output format clearly enough for an agent to follow.

1. 1.Copy the SKILL.md text below, or download the raw file.
2. 2.Create a folder named exactly voice-of-customer-scan in your agent's skills directory.
3. 3.Save the file inside that folder as SKILL.md.
4. 4.Ask the agent one of the trigger requests below.
5. 5.Check the output against what you already know before it leaves your desk.

• what are buyers saying
• voice of customer
• messaging from calls

---
name: voice-of-customer-scan
description: Pull buyer language, pain themes, objections, competitor mentions, product feedback, and how reps are pitching from one call or a folder of calls, with cited quotes, plus content seeds (LinkedIn hook, email, ad line) written in the buyer's own words. Use when PMM, marketing, product, or competitive asks "what are buyers saying", "voice of customer", "messaging from calls", "competitor mentions", "feature requests from calls", "objections we keep hearing", or "how are reps pitching this".
---

# Voice of Customer Scan

For PMM, marketing, product, and competitive intel. The best messaging is already in your calls, in words buyers use about their own problems. This skill finds those words, counts them, and cites them.

Rebuilt from the Momentum prompt library (the product executive brief across calls: Pain Points, Competitors, Product Feedback, Pitch, Objections; and Buyer Sentiment Content Shaper).

## Inputs

- One or more `ledger.md` files. For a folder, run `folder-recipe-run` first or read every ledger in `calls/`.
- `gtm-context.md` for product, competitors, and how buyers name them.
- Optional focus: a product, feature, or competitor the user names. If given, keep only rows that touch it.

## Step 1: Build the phrase bank

From PAIN, IMPACT, and OBJECTION rows on the buyer side, list the exact phrases buyers use for each problem. Group phrases that mean the same thing under one theme. Keep the buyer's words as the theme label when one phrase dominates.

| Theme | Buyer phrases (verbatim) | Calls | Accounts |

Count calls and accounts, not mentions. One loud account is one account.

## Step 2: Five sections

For each section: 2 to 4 themes, each with 2 or 3 bullets and one short paragraph. One quote per section maximum, the one that best proves the main theme. Never quote the same meeting twice in the report.

1. **Pain points**: voice of the customer only, never the rep.
2. **Competitors and alternatives**: who comes up, in what context (incumbent, evaluating, replaced, build it ourselves), and what buyers say about them. Include do-nothing.
3. **Product feedback**: requests, gaps, praise. Table:

| Request or feedback | Quote | Account | Meeting, time | Severity (blocker, important, nice to have) | Calls mentioning |

4. **Pitch**: how reps describe the product, what they lead with, which proof they use, and whether it matches the buyer's language from section 1. Mismatches are the finding.
5. **Objections**: the concern, how often, how reps answered, and whether the buyer accepted the answer (look for the next buyer line).

Quote citation format: `[Account - Meeting - Timestamp]`. Trim filler words from quotes; never change meaning.

## Step 3: Content seeds

Written from the phrase bank, in the buyer's words. Address at least 3 specific pains. No generic marketing language.

- **LinkedIn hook** (2 lines): opens with a buyer phrase.
- **Email opener** (3 sentences): signal, pain in their words, one question.
- **Ad line** (under 15 words): the top pain phrase plus the outcome.

Never put a customer's name or identifying detail in a seed. Label seeds "draft, needs approval before use".

## Output: `voc.md`

```markdown
# Voice of Customer: <scope>, <date range>
Calls: <n> | Accounts: <n> | Focus: <product or none>

## Phrase bank
## Pain points
## Competitors and alternatives
## Product feedback
## Pitch vs buyer language
## Objections
## Content seeds (draft)
## Method
Calls read, date range, rows used, what was excluded.
```

Keep the report under 8,000 characters, sections under 2,000 each.

## Rules

- Buyer words only in Pain points and the phrase bank.
- Count accounts, not mentions, before calling something a theme. A theme needs 2 or more accounts; a single-account item goes under "Watch".
- Every quote cited. No quotes invented or merged from two speakers.

What does the Voice of Customer Scan skill do?

Pull buyer language, pain themes, objections, competitor mentions, product feedback, and how reps are pitching from one call or a folder of calls, with cited quotes, plus content seeds (LinkedIn hook, email, ad line) written in the buyer's own words.

Where does the Voice of Customer Scan skill come from?

Built by Jonathan Kvarfordt on the GTM AI OS. Supplied as a complete zip and served unchanged.. The source and included files are shown on this page and in the package download.

Why was the Voice of Customer Scan skill chosen for this library?

It covers a job one of the eight revenue seats has to finish, and it states its process, decision rules, and output format clearly enough for an agent to follow.

When should the Voice of Customer Scan skill not be used?

Use it only for the job named above. A skill applied outside its scope produces confident output on the wrong question.

How do I install the Voice of Customer Scan SKILL.md file?

Download the file, create a folder named exactly voice-of-customer-scan inside your agent's skills directory, and save the file inside it as SKILL.md. The agent loads it when a request matches the description.