Podcast Analytics: What to Measure (and What to Ignore)

Downloads are a vanity metric. The signals that actually predict a healthy, growing show.

The landscape

When it comes to cutting through vanity metrics to the signals that matter, the details matter. Here's what the evidence and practitioner experience point to:

• Downloads measure reach; completion rate measures quality. A show retaining 80%+ of listeners per episode will compound; one at 30% won't.
• Follower/subscriber counts on Spotify/Apple predict future downloads better than any single episode's numbers — track the trend, not the noise.

What actually works

When it comes to cutting through vanity metrics to the signals that matter, the details matter. Here's what the evidence and practitioner experience point to:

• Drop-off points within episodes are diagnostic gold — consistent exits at the same timestamp signal a structural problem (a bad intro, a boring segment).
• Listening apps report per-listener engagement: repeat listeners vs one-off downloads shows habit formation — the real growth signal.

Putting it into practice

When it comes to cutting through vanity metrics to the signals that matter, the details matter. Here's what the evidence and practitioner experience point to:

• Attribution is inherently weak in audio; use promo codes, unique links and 'how did you hear about us' surveys to close the loop.
• Review analytics monthly, not daily — podcast data is noisy; weekly deltas mean nothing, monthly trends mean everything.

Key takeaways

• Downloads measure reach; completion rate measures quality. A show retaining 80%+ of listeners per episode will compound; one at 30% won't.
• Listening apps report per-listener engagement: repeat listeners vs one-off downloads shows habit formation — the real growth signal.
• Review analytics monthly, not daily — podcast data is noisy; weekly deltas mean nothing, monthly trends mean everything.

Ready to put this into practice? PodKrafter turns research into broadcast-ready podcast episodes — start free.