Podcast Appearances (Contact)
Podcast appearances by individual contacts—what they said, the topics they care about, and where they're building a public voice.
Overview
Podcast Appearances (Contact) capture the moments a prospect goes on the record—what an executive actually said on a podcast, in their own words, attributed to the specific person speaking.
We process episodes across a large catalog of business and industry shows, transcribe them, and extract structured signals tied to the individual contact. Each row carries one of 43 subtypes, the supporting quotes, and our estimate of when the event took place.
There are two row shapes, distinguished by signal_subtype:
- Appearance rows —
signal_subtype: "podcastAppearance". One row per guest per episode: the host, the topics covered, the occasion, their key talking points (each with a supporting quote), co-guests, and a snapshot of what their company does. - Claim rows — any other subtype (e.g.
aiInvestment,fundingEvent,techAdoption,careerHistory). A single quote-backed claim the contact made, usually about their own company. Claims about other companies are delivered as Podcast Mentions (Company).
The data object differs between the two—see Data Object below.
Available Subtypes
The signal_subtype field takes one of the 43 values below. The group headings are an organizational grouping for readability only—the group is not a field in the payload. Rows delivered is the number of rows delivered from May 31 to Sep 20, 2026; it shows which subtypes are common and which are rare.
podcastAppearance and careerHistory exist only on this contact type. podcastSponsorship exists only on Podcast Mentions (Company).
Appearance (1)
One row for each guest on each episode.
| Subtype Enum | Description | Rows delivered |
|---|---|---|
podcastAppearance | Person appeared as a guest on a podcast episode | 20,948 |
Themes (20)
Priorities and initiatives the speaker voices. The tokens match the theme subtypes on earnings-call and SEC signals.
| Subtype Enum | Description | Rows delivered |
|---|---|---|
aiInvestment | Company invests in AI models, tools or AI features | 2,353 |
dataInvestment | Company invests in data infrastructure, analytics or data quality | 817 |
cybersecurityInvestment | Company invests in security tools, teams or programs | 151 |
cloudInvestment | Company invests in cloud infrastructure or a cloud migration | 12 |
automationInvestment | Company invests in process or workflow automation | 496 |
digitalTransformation | Company runs a broad digital-transformation program | 371 |
legacyModernization | Company replaces or modernizes legacy systems | 159 |
techInvestment | Company invests in technology that no narrower theme covers | 873 |
customerExperienceFocus | Company makes customer experience a stated priority | 1,557 |
efficiencyFocus | Company makes operational efficiency or productivity a stated priority | 930 |
costReduction | Company cuts costs or runs a cost-reduction program | 151 |
marketExpansion | Company moves into new markets, segments or verticals | 2,408 |
internationalGrowth | Company expands into new countries or regions | 219 |
platformStrategy | Company builds a platform or ecosystem strategy | 1,185 |
strategyShift | Company changes direction, business model or focus | 1,943 |
talentChallenge | Company struggles to hire or keep talent | 226 |
supplyChainPain | Company has supply-chain or sourcing problems | 11 |
scalingChallenge | Company struggles to scale operations, systems or teams | 44 |
complianceBurden | Company carries a regulatory or compliance burden | 92 |
strategicInitiative | Company names a strategic initiative or company-level priority | 2,854 |
Knowledge (8)
Facts about the stack, buying behavior and opinions.
| Subtype Enum | Description | Rows delivered |
|---|---|---|
painPointDisclosed | Operational pain voiced by someone at the company | 2,770 |
techAdoption | Named tool or vendor currently in the stack | 3,073 |
buyingIntent | Company is actively evaluating a solution now | 76 |
buyingCriteria | How they select vendors (build-vs-buy, budget rules) | 1,134 |
churnRisk | Dissatisfaction with, or migration away from, a named vendor | 40 |
executiveOpinion | A stance strong enough to warrant a row | 260 |
industryPrediction | Investor or analyst thesis about the company or its market | 506 |
hiringSignal | Hiring plans or hiring challenges at the company | 806 |
Events (11)
First-party disclosures: the contact states an event at their own company.
| Subtype Enum | Description | Rows delivered |
|---|---|---|
fundingEvent | Company raised or is raising a funding round | 421 |
acquisitionAnnounced | Company announced an agreement or plan to acquire another company | 71 |
acquisitionCompleted | Company completed an acquisition | 326 |
productLaunch | Company launched or will launch a product or service | 2,195 |
partnership | Company has a partnership or alliance with a named company | 1,373 |
expansion | Company adds offices, facilities, locations or capacity | 807 |
leadershipChange | Company has a new executive or an executive departure | 307 |
layoffs | Company reduced or will reduce headcount | 18 |
majorContractWin | Company won a large contract or customer | 304 |
cybersecurityIncident | Company had a breach or other security incident | 5 |
achievesCertification | Company earned a certification or compliance standard (for example SOC 2 or ISO) | 100 |
Durable history (3)
Past facts that stay useful. The subtype name marks them as old, so you opt in with a filter.
| Subtype Enum | Description | Rows delivered |
|---|---|---|
careerHistory | Guest's track record: exits, companies founded or sold, well-known past clients | 10,560 |
formerVendor | Company used a named vendor and left (displacement intel) | 375 |
formerClient | Company previously served a named client | 66 |
Schema — Appearance row (podcastAppearance)
podcastAppearance){
"signal_id": "b47a9eb9-2b2f-5e75-865a-03409f2effa2",
"batch_id": "2026-09-20",
"signal_type": "podcast-contact",
"signal_subtype": "podcastAppearance",
"signal_name": "Tom McKenna (Global Head of Media Operations and Business Development @ Audible) on Strictly Business",
"detected_at": "2026-09-20T15:05:23Z",
"association": "contact",
"company": {
"name": "Audible",
"domain": "audible.com",
"linkedin_url": "linkedin.com/company/audible",
"industries": [
"Software Development"
],
"employee_count_low": null,
"employee_count_high": null,
"description": "Audible provides a platform for accessing audio-based content."
},
"contact": {
"full_name": "Tom McKenna",
"first_name": "Tom",
"last_name": "Mckenna",
"job_title": "Svp, Global Head of Media, Ops & Business Development (L8)",
"email": "[email protected]",
"linkedin_url": "https://www.linkedin.com/in/mckennatom"
},
"data": {
"episode_id": "60588532285",
"podcast_name": "Strictly Business",
"episode_title": "The Future of Media Buying: Turning Buzz into Business With WPP Media’s Nancy Hall and Audible’s Tom McKenna",
"episode_url": "https://omny.fm/shows/strictly-business-1/the-future-of-media-buying-turning-buzz-into-business-with-wpp-media-s-nancy-hall-and-audible-s-tom-mckenna",
"transcript_source": "transcript",
"headline": "Tom McKenna (Global Head of Media Operations and Business Development @ Audible) on Strictly Business",
"evidence": null,
"is_ad": false,
"published_at": "2026-09-18",
"recorded_at_estimate": "2026-09-17",
"date_confidence": "exact",
"occasion": "Speaking at Variety's Entertainment and Technology Summit",
"host": "Cynthia Littleton",
"company_snapshot": {
"what_they_do": "Audiobooks and audio entertainment",
"org_facts": [],
"technologies_used": [
"Dolby Atmos"
],
"customers_mentioned": [
"British Airways",
"JetBlue",
"Twitch",
"NBC Universal"
],
"competitors_mentioned": []
},
"topics": [
"Media strategy",
"Content formats",
"Community engagement",
"Brand campaigns"
],
"talking_points": [
{
"point": "Full-cast, cinematic audio productions are performing well.",
"quote": "full cast cinematic audio productions tend to work really, really well. We're seeing that resonate. It about the immersive sound design It about the Dolby Atmos the 360 degree sound We seeing that work really really well."
},
{
"point": "Creator-led storytelling provides authentic promotion.",
"quote": "We also seeing creator storytelling tend to work really really well for us So think about like hearing it naturally in a voice... it comes from them, their own voice, their own fan and their own influencer group."
},
{
"point": "Building fandom requires creating exclusive, behind-the-scenes access.",
"quote": "there's an insider status that I think starts to become really popular here. You've got to do that. Behind the scenes, clips, exclusive content. So making of videos tends to work really, really well."
}
]
}
}Schema — Claim row (e.g. productLaunch)
productLaunch){
"signal_id": "7cdccdd0-c203-5147-b57e-693051dee800",
"batch_id": "2026-09-20",
"signal_type": "podcast-contact",
"signal_subtype": "productLaunch",
"signal_name": "Brex · productLaunch · completed (2026-09)",
"detected_at": "2026-09-20T15:05:33Z",
"association": "contact",
"company": {
"name": "Brex",
"domain": "brex.com",
"linkedin_url": "linkedin.com/company/brexhq",
"industries": [
"Financial Services"
],
"employee_count_low": null,
"employee_count_high": null,
"description": "Brex provides a financial platform that integrates corporate charge cards, cash management accounts, and expense and travel management tools."
},
"contact": {
"full_name": "Pedro Franceschi",
"first_name": "Pedro",
"last_name": "Franceschi",
"job_title": "Founder and Chief Executive Officer",
"email": "[email protected]",
"linkedin_url": "https://www.linkedin.com/in/pfranceschi"
},
"data": {
"episode_id": "60341295305",
"podcast_name": "Behind the Craft",
"episode_title": "Stop Building AI Agents. Build AI Employees Instead (Live Demo) | Pedro Franceschi",
"episode_url": "https://podcasters.spotify.com/pod/show/peter-yang42/episodes/Stop-Building-AI-Agents--Build-AI-Employees-Instead-Live-Demo--Pedro-Franceschi-e3ond3v",
"transcript_source": "transcript",
"headline": "Brex · productLaunch · completed (2026-09)",
"evidence": "that's where we build Crab Trap. So the idea of Crab Trap and actually I can just show you the GitHub repo. Yeah, Crab Trap is open source, right? So any company can do it. Yeah, so Crab Trap is open source and really what it does is it's like proxy that intercepts requests.",
"is_ad": false,
"published_at": "2026-09-13",
"recorded_at_estimate": "2026-09",
"date_confidence": "inferred",
"quotes": [
"that's where we build Crab Trap. So the idea of Crab Trap and actually I can just show you the GitHub repo. Yeah, Crab Trap is open source, right? So any company can do it. Yeah, so Crab Trap is open source and really what it does is it's like proxy that intercepts requests."
],
"speaker": "Pedro Franceschi",
"speaker_relation": "own_company",
"timing": "completed",
"event_date": "2026-09"
}
}Field Reference
Core Fields
| Field | Type | Required | Description |
|---|---|---|---|
signal_id | string (UUID) | ✓ | Unique, deterministic identifier for this signal (the same fact always gets the same ID) |
batch_id | string (date) | Weekly processing batch that produced the row (YYYY-MM-DD). Present in GCS bucket files; API responses can omit it | |
signal_type | string | ✓ | Always "podcast-contact" |
signal_subtype | string | ✓ | podcastAppearance or a claim subtype (see allowed values above) |
signal_name | string | ✓ | Human-readable label (same value as data.headline). Appearance rows: Name (Title @ Company) on Show. Claim rows: Company · subtype · timing (event_date) |
detected_at | string (ISO 8601) | ✓ | When Autobound detected/ingested the signal (not the episode date—see data.published_at) |
association | string | ✓ | Entity association type—always "contact" |
company | object | ✓ | The contact's company (see Company Object) |
contact | object | ✓ | The individual the signal is about (see Contact Object) |
episode_idis nested underdata, not at the top level. There is noparent_signal_idfield.
Company Object
| Field | Type | Required | Description |
|---|---|---|---|
company.name | string | ✓ | The contact's company |
company.domain | string | ✓ | Company domain (primary match key) |
company.linkedin_url | string | Company LinkedIn URL without scheme (e.g. linkedin.com/company/brexhq); nullable | |
company.industries | array[string] | Industry classifications; nullable | |
company.employee_count_low | integer | Lower bound of employee-count range; nullable | |
company.employee_count_high | integer | Upper bound of employee-count range; nullable | |
company.description | string | Company description; nullable |
Contact Object
| Field | Type | Required | Description |
|---|---|---|---|
contact.full_name | string | ✓ | Full name of the contact |
contact.first_name | string | First name | |
contact.last_name | string | Last name; nullable | |
contact.job_title | string | Job title on record; nullable. It can differ from the title stated on the episode, which appears in signal_name on appearance rows | |
contact.email | string | Work email (match key); nullable | |
contact.linkedin_url | string (URL) | ✓ | Contact LinkedIn profile URL (match key) |
Data Object
Fields on every row:
| Field | Type | Required | Description |
|---|---|---|---|
data.headline | string | ✓ | One-line summary of the signal (same value as signal_name) |
data.is_ad | boolean | ✓ | Always false on contact rows (sponsorships are company-only) |
data.podcast_name | string | ✓ | Name of the podcast show |
data.episode_title | string | ✓ | Title of the episode |
data.episode_url | string (URL) | Link to the episode; nullable | |
data.episode_id | string | ✓ | Source podcast episode identifier |
data.published_at | string (date) | ✓ | Episode publication date (YYYY-MM-DD) |
data.recorded_at_estimate | string | ✓ | Best estimate of when the episode was recorded (YYYY-MM or YYYY-MM-DD) |
data.date_confidence | string | ✓ | Confidence in the date fields: exact, inferred or unknown |
data.transcript_source | string | ✓ | Origin of the analyzed text (currently always transcript) |
data.evidence | string | Primary supporting quote (the first entry of quotes); always null on appearance rows |
Fields present only on appearance rows (podcastAppearance). Each can be absent when the episode does not supply it:
| Field | Type | Description |
|---|---|---|
data.host | string | Podcast host name |
data.occasion | string | The occasion / context of the appearance |
data.co_guests | array[object] | Other guests on the same episode |
data.co_guests[].name | string | Co-guest name |
data.co_guests[].title | string | Co-guest title as stated on the episode |
data.co_guests[].employer | string | Co-guest employer as stated on the episode |
data.topics | array[string] | High-level topics discussed in the episode |
data.talking_points | array[object] | Key points the contact made (see below) |
data.talking_points[].point | string | Paraphrased key point |
data.talking_points[].quote | string | Supporting verbatim quote |
data.company_snapshot | object | Context about the contact's company from the episode |
data.company_snapshot.what_they_do | string | What the company does, in the speaker's words (nullable) |
data.company_snapshot.org_facts | array[string] | Stated numbers and facts about the company |
data.company_snapshot.technologies_used | array[string] | Tools and vendors mentioned as in use |
data.company_snapshot.customers_mentioned | array[string] | Customers named on the episode |
data.company_snapshot.competitors_mentioned | array[string] | Competitors named on the episode |
Fields present only on claim rows (every subtype except podcastAppearance):
| Field | Type | Description |
|---|---|---|
data.speaker | string | Name of the person who made the statement |
data.speaker_relation | string | Speaker's relationship to the company: own_company (the usual value) or third_party; can be absent |
data.quotes | array[string] | Verbatim transcript quotes supporting the signal |
data.timing | string | When the described event occurs relative to the recording: completed, in_progress, imminent (about 60 days), planned, aspirational or ongoing_state (a standing fact, not an event) |
data.event_date | string | Best estimate of the event date (YYYY, YYYY-MM or YYYY-MM-DD); for ongoing_state it is the as-of date. Can be absent |
data.event_date_text | string | Raw temporal phrase from the transcript (e.g. "October 9th"), when present |
Coverage
- Bucket:
gs://autobound-podcast-contact/ - Refresh: Weekly
- Association: Contact-level
- Subtypes: 43 (
podcastAppearance+ 42 claim subtypes)
Updated 12 days ago

