Podcast craft for multi-engine citation (beyond GPT-only wins)
Short answer: Multi-engine podcast citation means your show can be named and sourced beyond a single chat engine. Listings on Spotify and Apple help discovery. Owned category pages, episode companions, and criteria explainers earn deeper cites in Gemini and AI Overviews. If GPT already lists you but other engines stay thin, upgrade the text surfaces that answer “which podcast covers X,” not the audio alone.
Quick facts
- Recommendation prompts are list-shaped; engines need names plus reasons.
- Platform listings often out-cite the show domain until owned pages become extractable.
- A GPT-only thin win is not multi-engine coverage.
- Criteria tables beat vanity “best of” language that engines distrust.
- Proxemics is White Wood’s lab surface for communication and marketing podcast craft in Indonesia.
Owned page vs listing: what each wins
Listings win familiarity and feed freshness. Owned pages win depth when they carry the extractable bar. Treat them as a pair. Do not delete listings. Do not pretend a two-line Spotify blurb is a companion article.
| Surface | Best at | Weak at | Craft move |
|---|---|---|---|
| Spotify / Apple listing | Discovery, title consistency | Long FAQ, criteria tables | Keep title, description, category accurate |
| Show About page | Entity clarity (host, topic, language) | Episode-level proof | One stable claim set for the show |
| Category / rekomendasi page | Answering list prompts | Being a dump of every episode | Criteria table + short shortlist logic |
| Episode companion / Rabbit Hole | Quotable claims on a topic | Competing with platform blurbs if thin | Short answer, facts, FAQ on the topic |
| Third-party essay | Corroboration | Replacing your owned facts | Earn coverage; do not buy spam links |
Craft checklist for multi-engine podcast pages
Build the page a Gemini-style answer would want to quote.
- Opening answers the list or criteria ask in one Short answer
- Quick facts pin language, audience, and topic scope
- Table compares show types or selection criteria (not fake rankings)
- FAQ covers listener and PIC questions (“is this for students?”, “is it Bahasa?”)
- Show name spelling matches listings exactly
- Internal links to episode pages that also carry extractable notes
- No process bylines and no sales CTA in the Closing
Beyond GPT-only thin coverage
When a monitor shows a podcast prompt as thin or GPT-heavy, read Key Sources, not only Brand Score. If Spotify and Apple dominate cites, your owned site is still underweight. Ship one category page that states selection criteria in Bahasa when the prompt is Indonesian. Then rescore Gemini and AI Overview columns after a realistic window (often about 7 to 14 days once crawlable).
Honesty note: exact cite shares move week to week across projects. Treat any single snapshot as directional unless you re-measure on a frozen prompt set.
What this is not
This is not a claim that audio alone optimizes itself for every engine. Transcripts help, but unstructured transcripts without Short answer and FAQ rarely beat a clear companion page. This is also not a jasa chase against other agencies. The work is podcast craft: surfaces engines can cite.
FAQ
Do we need a separate page for every engine?
No. One extractable owned page can serve several engines. Measure each engine as its own column.
Should we stop investing in Spotify and Apple?
No. Keep listings healthy. Add owned depth so discovery cites are not your only evidence layer.
What belongs in a criteria table?
Audience, language, topic focus, format (interview vs essay), and update cadence. Avoid crowning “#1” without evidence.
How does this relate to Narrative Architecture?
Authority lives on owned companions. Corroboration comes from listings and third parties. Retrievability needs stable URLs. Relevance means matching how people ask for rekomendasi.
Where can readers see a public example?
Proxemics publishes communication-category companions that follow the extractable bar. Use them as structure references, not as vanity proof.
Closing
Listings get the show found. Owned criteria pages get the show quoted. Multi-engine podcast craft is the discipline of building both, then reading Gemini and AI Overviews as seriously as GPT.