Build a voice-of-customer database and content vaults

Coming

Ingests sales calls, lessons, testimonials, YouTube transcripts and comments, Facebook comments, DMs and email replies into an immutable raw store, extracts verbatim quotes tagged by funnel stage and use case, builds theme-indexed voice banks, and supports fresh-angle re-mining. Includes the Fathom and Vimeo bulk ingest with quote audit and repair.

Kit

content-tools

You need

Account	Why	Free or paid	Plan
Fathom (call recorder)	Client-side call recorder used by clients for lessons and sales calls. The fathom-download skill pulls the MP4 and speaker transcript from a share link; the voc-ingest scripts pull whole libraries through the API.	Free tier available	Free for recording and sharing; the API and webhooks need Fathom Premium or Team.
Fireflies.ai	Records and transcribes Zoom and Google Meet calls; the GraphQL API pulls transcripts into the voice-of-customer database and the Brain lessons.	Free tier available	Free plan gives limited transcription credits; Pro (about 10 USD/mo) for unlimited transcripts and API access. API access needs a paid plan.
Vimeo	Hosts a client's lesson and webinar library; voc-ingest downloads the VTT captions for quote mining.	Free tier available	Whatever plan the client already has; captions and the API are available on paid plans (Starter and up).
yt-dlp	Downloads YouTube, TikTok and Instagram video, auto-captions and comments for vaults, B-roll, MIDI extraction and quote verification.	Free	Free.
Python 3.12 + uv	Runs every pipeline script (video gates, library export, research cache, Brain export, PDF generator, MIDI tools) inside per-project virtual environments.	Free	Free.
Meta Business Suite (Pixel, Conversions API, Marketing API)	Facebook and Instagram ads. The funnel fires Pixel events, the attribution system sends server-side Conversions API events, the Meta Ads tab pulls ad-level insights, and social posting reaches your Pages through Zernio.	Free	Business account is free; you pay for ad spend only. Requires a verified business and a domain verification TXT record for the Pixel.
HYROS	Paid attribution for a client running ads; the Channel Attribution tab proxies the HYROS API. Timo built a first-party clone so new clients can skip it.	Paid	Paid only, priced on ad spend (hundreds of dollars a month). Only needed if the client already runs it; otherwise use the first-party attribution system.
GitHub	Every site, dashboard and workflow lives in a repo. Actions run scheduled jobs (email snapshots, daily scrapes). Issues double as an email alert channel for bookings.	Free	Free plan. Actions minutes on the free plan cover the cron jobs used here (a few minutes a day).
YouTube Data API v3 (Google Cloud)	Exact per-video view counts for the $ per view tab, channel stats, and (with OAuth) uploads. Public metadata, captions and comments come through yt-dlp without a key.	Free	Free; 10,000 quota units per day is plenty for daily syncs.
Slack (incoming webhooks)	Client review pings from the dashboard (script approved, email needs changes) and social-post notifications.	Free	Free workspace is enough for webhooks.
Supabase	Postgres database, auth (dashboard logins), row level security, Realtime pulse, Edge Functions and pg_cron for every dashboard, the attribution tables, booking alerts and the Brain state.	Free tier available	Pro (25 USD/mo) for anything a client logs into. The free tier pauses a project after 7 days without traffic and caps an org at 2 active projects; a paused free project took three client dashboards offline at once (a project log).

What you bring

Your call recordings or transcripts, exported
Your testimonials, and written permission for each one
Your label on every person in those calls: customer or prospect, never left blank
Your video links with your own verdict on each: winner or flop, and the sales count
Your ad exports from Ads Manager, and your ad account ids
Your YouTube channel, by handle, plus your comments
Your repository secrets for the deploy

The steps

Bring your materials
Install the tools and initialise the database
Ingest the raw material (immutable)
Extract verbatim quotes in parallel batches
Mine fresh angles and build the voice bank
Store one YouTube video in the video vault
Store one ad in the ad vault
Automate the daily channel scrape and the weekly sentinel
Prove the corpus is trustworthy before any engine reads it

Get the full playbook

Plug in your AI Read it in the Library

Part of: Get your database sorted