GTM SCIENCE

What this measures: of every session that arrived on usc.com from a podcast UTM link, how many landed first on a high-intent page — /frameworks/…, /newsletter/…, /pricing/, or /howwework/. These pages signal a real evaluation step (downloading a lead magnet, checking pricing, exploring services), not casual browsing. Timeframe: trailing 90 days from GA4. Why it matters: raw "sessions from the podcast" only tells you traffic showed up. Intent-page rate tells you whether the listener was a prospect when they arrived — the gap between an interested ear and a qualified lead.

What this measures: for each Spotify episode released inside the last 90 days, compares average daily sessions in the 7 days after release to the 30 days before. Because GTM Science releases roughly weekly, the baseline includes other release weeks — so the metric reads as "did this release outperform a typical release week?" rather than vs a hypothetical no-release baseline. Captures untagged-but-real podcast lift (listeners who typed the URL or searched the brand) that pure UTM tracking misses. Limitation: correlation, not causation — coincident traffic from email, ads, or organic search can be misread as podcast lift. Episodes outside the 90-day window or without 7+ days of baseline data are skipped.

What this measures: hygiene check on episode descriptions — does each episode's show-notes link to usc.com use the expected UTM convention (utm_source=podcast&utm_medium=podcast-shownotes&utm_campaign=<slug>)? Scans YouTube descriptions live and the Spotify-hosted RSS feed (which also serves Apple Podcasts). Drift = both platforms tag the CTA but use different campaign slugs — the worst class of attribution problem because traffic can't be reconciled across platforms.