Insights
There's been a lot of mystery in the Podcast SEO world for years: how do podcast app search engines actually work?
In July 2021, PodNews published an article outlining what Apple Podcasts indexes for search. At that time, the key fields were:
My independent testing in June 2025 confirms that this still holds true—with one significant addition: transcripts.
Between 2021 and now, AI transcription has become so affordable that Apple now provides transcripts for free to all podcasters (in supported languages). This is a welcome improvement for accessibility and discoverability.
While developing PodKeyScore.com, I noticed one podcast was ranking for a phrase that didn't appear in its metadata. I checked the Apple-generated transcript and found the exact phrase there. That was the confirmation I needed.
“The Apple Podcasts search engine is using the contents of their generated transcripts as part of the index.”
To verify it wasn’t a one-off, I tested podcasts with few ratings and limited metadata. I searched for unique transcript phrases that didn’t appear in titles or descriptions. Results consistently showed those podcasts ranking for transcript-based terms.
There are still open questions. For example, does Apple index self-submitted transcripts? My suspicion is no, likely to prevent keyword stuffing abuse.
Regardless, this shift is great news. Instead of stuffing metadata, creators can now focus on making valuable audio content. And with PodKeyScore, it's easier than ever to discover what keywords your show ranks for—and which to target next.