Every Place to Inventory Your Podcast So LLMs Cite You (and Why the Ahrefs Numbers Lie)

When we inventoried 700+ podcast appearances across 5 platforms, the platforms disagreed on which episodes existed — about 35% were findable on only ONE platform. That article covered the big five. This one is the complete map: every place a podcast (or a podcast guest) should be inventoried, and the honest math on what it earns you — because the payoff shows up somewhere most SEO tools never look.

The complete inventory map

Tier 1 — The discovery core (do these first)

• Podchaser — the IMDb of podcasts. Claim your creator profile (verification runs through X), confirm every credit individually, add socials + site. Wikidata property: P9743.
• Listen Notes — crawls RSS and indexes transcripts, so it finds episodes where your name was spoken but never titled. Export the CSV; turn on name alerts. Wikidata property: P10213.
• Apple Podcasts — title-keyword search only, but the directory every other app syncs from. Claim your show in Podcasts Connect.
• Spotify — claim in Spotify for Creators; video episodes count double now.
• YouTube — the platform audio-only directories miss entirely. Inventory it separately; request collaborator credits on guest appearances.

Tier 2 — The long-tail directories (each catches strays)

Podcast Index (the open database many AI crawlers read directly) · OwlTail (84 entries for Dennis; a handful existed nowhere else) · Player FM · Podtail · Podbay · Castbox · TuneIn · iHeartRadio · Amazon Music / Audible · Deezer · Pocket Casts · Goodpods · Podcast Addict. You don’t maintain profiles on all of these — you verify your RSS feed reached them (submit once where it hasn’t) and record what each shows in your master sheet. Their pages become part of your citation surface whether or not humans browse them.

Tier 3 — The entity layer (where citations become machine-readable)

• Your own /podcast-appearances/ page — the hub. Model: 619 Dennis Yu podcast appearances, with the count in the H1, the full table, and a “how we built this” methodology section — which is exactly the part LLMs quote.
• Wikidata — add P9743 (Podchaser creator ID) to the person, P10213 (Listen Notes ID) to the show, referenced to the live profiles. This is the structured layer Google’s Knowledge Graph and every major LLM ingest. Our Wikidata SOP covers the mechanics.
• Grokipedia — check your article; suggest corrections citing your appearances page and claimed profiles. Grok already cites BlitzMetrics pages more than any other AI platform (data below).
• Google Knowledge Panel — claimed podcast profiles are corroboration; wire them in per our Knowledge Panel guide.
• MuckRack / press profiles — journalists’ databases index podcast appearances as press credits.

The booking layer (bonus)

MatchMaker.fm, PodMatch, and Rephonic — where hosts research guests. A profile citing your verified credit count converts here; “608 credited episodes” is a booking argument no bio paragraph can match.

Why the Ahrefs numbers look like nothing happened

Both panels are real Ahrefs data pulled the same day. The left panel is why smart people give up on citation work: the organic view samples a keyword database, so it structurally undercounts zero-volume long-tail queries (the “[your name] + [show name]” searches that podcast inventories win), it ignores retrieval entirely, and it can’t see inside an AI answer. The right panel is the same toolset pointed at the surface that actually matters now — and it shows 67 citations of blitzmetrics.com inside AI-generated answers, with Grok leading at 17. Directory links are mostly nofollow and the profile pages are thin, so classic link metrics barely move. That was never the mechanism. The mechanism is consistency of the entity across many trusted, structured sources — which is what retrieval systems select for when they decide who is real.

Three honest caveats, so this stays a measurement article and not a hype piece: (1) AI-citation tracking is itself a sample of prompts — treat direction, not decimals. (2) The pages in this cluster are months old with zero referring domains; Google rankings should be near zero at this stage, and the fix (internal links, nav placement, first backlinks) is ordinary SEO. (3) Google Search Console is the ground truth for long-tail impressions — check it before believing any third-party keyword count, ours included.

Case study: what this looked like for Dennis, in one day

• Verified the live Podchaser profile: 619 credits across 353 podcasts (11 new since the March audit, queued for individual triage — never bulk-confirm).
• Fixed the master list: dated update block, corrected counts, repaired table, canonical links, reciprocal link to the scored tracker.
• Put the appearance count and master-list link into the author bio that renders on every post sitewide — hundreds of internal links from one edit.
• Packaged the whole workflow as a downloadable Claude skill: podcast-appearance-syndicator.
• Scheduled a weekly agent that re-checks the live counts, syncs the pages, and logs the AI-citation numbers — so this article’s data ages into a growth curve instead of a snapshot.

Do this for yourself: install the skill file, then follow the agent activation checklist — access, scheduled jobs, audit, goals. If you’d rather we run it with you, that’s what the team does every week for clients.

Related resources

The definitive podcast inventory guide · Podchaser SOP · Listen Notes SOP · Finding what aggregators miss · The Definitive Guide to Podcasting · The Task Library