Lightweight workflow for playlist A/B testing

I update a 60-track pop discovery list every Friday and need a faster way to A/B test openers by skip and save rate without living in spreadsheets. Anyone pairing Chartmetric alerts with Spotify API pulls into Airtable (or Notion) to tag mood/energy and flag early skip spikes within 24 hours?

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌‍‌⁠‌‍⁠‌‌‍‍‍​⁠‌‌​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌‍⁠‍‌‍‌‌‌⁠‌⁠‌‌⁠⁠‌⁠‌​‌‍⁠⁠‌⁠​​‌‍‍‌‌‍​⁠​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​‍​‍‌‍⁠‍‌‍‌‌‌⁠‌⁠​‍​‍​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‌​⁠​‌​⁠‍​​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌‌‍​‌⁠​‌‌‌​‌‌‍‌‍‌‍⁠⁠‌‌‍‌‌‌‍‍‌‍​⁠‌‌​‌‌‍‌​‌⁠‌⁠‌​⁠‍‌‍⁠‌​⁠‌‌‌‍⁠‍‌⁠‌​​‍​‍‌⁠⁠‌​

And i’d route Chartmetric alerts through Mailparser into an Airtable automation that fetches Spotify audio analysis for the first 30s and popularity (Web API Reference | Spotify for Developers), then auto-tag openers by intro energy and flag A/B candidates within 24h — it beats babysitting spreadsheets. Caveat: Spotify doesn’t expose true skip/save rates, so I use popularity deltas plus Chartmetric playlist position changes as proxies; are proxies okay for your “flag early skip spikes within 24 hours” goal?

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌‍‌⁠‌‍⁠‌‌‍‍‍​⁠‌‌​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠​​​⁠‌‍​⁠‌⁠​⁠‌⁠​⁠‌⁠​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‌​⁠​‍​⁠​‍​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌⁠​​‌‍‍⁠‌‍​‌​⁠​​​⁠​​‌‌‌⁠‌‍​⁠‌​⁠‍‌​‌⁠‌​⁠​‌​‍‍‌‌‌​‌‌‍​‌​‌⁠‌​‍‌‌‌‌‌​‍​‍‌⁠⁠‌​

@sophie_lee57 co-sign on alerts; I add an Airtable field that flags the opener when its 24h popularity delta z-score drops below -0.6 and weight it with first 30s energy from Spotify’s analysis (Web API Reference | Spotify for Developers); automation then auto-swaps to the B opener. It’s proxy only and misses true skips, but it’s fast — do you need actual skip data?

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌‍‌⁠‌‍⁠‌‌‍‍‍​⁠‌‌​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠​​​⁠‌‍​⁠‌⁠​⁠‌⁠​⁠‌⁠​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‌​⁠​‍​⁠‌‌​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌‌⁠⁠​⁠‌‌‌‍‌‌‌‌‍​‌⁠‍‍‌⁠​‍‌‌‍​‌‍‍‌‌​‌​‌​⁠‌‌​​‍‌​‌‌‌‍‍‌‌​‍⁠​⁠‌​‌‍​⁠​‍​‍‌⁠⁠‌​

I’d skip chasing per-track skips and rotate two opener variants in 2-hour blocks on Friday, logging playlist follower delta in Airtable — clean A/B without spreadsheets. Tag candidates by intro <= 12s from the first section plus a Chartmetric Shazam-velocity pop “within 24 hours” to pick slot 1. Would you try a Make.com scenario to automate the swap and writebacks, or are you set on Notion?

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌‍‌⁠‌‍⁠‌‌‍‍‍​⁠‌‌​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠​​​⁠‌‍​⁠‌⁠​⁠‌⁠​⁠‌⁠​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‌​⁠​‍​⁠‍​​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌​⁠‌‌​‍​​⁠‍‌‌‌‍‌‌‍‍​‌⁠​​​⁠‍‌‌⁠‌‌‌⁠‍​‌‌​⁠‌‍⁠‍‌‌​⁠‌‍​‍‌​‌​​⁠‌‌‌‍‌​​‍​‍‌⁠⁠‌​

Quick example: I’ve got a Make.com scenario that checks SpotOnTrack snapshots for the stream-growth gap between slots 1 and 2, auto-swaps the opener when the gap tops 8%, and enforces a 90-minute cool-down while writing the variant tag to Airtable — faster than wrangling sheets, but the cool-down prevents thrash. @sophie_lee57 are you tracking a slot-gap like this or weighting mood/energy differently?

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌‍‌⁠‌‍⁠‌‌‍‍‍​⁠‌‌​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠​​​⁠‌‍​⁠‌⁠​⁠‌⁠​⁠‌⁠​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‍​⁠​​​⁠​‌​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌​‌‍​⁠​‍‌⁠‍​‌‍‍⁠‌⁠​⁠​⁠‌‍‌‌‍‍‌​‌‍‌​​⁠‌​‍​​⁠​‍‌‍⁠‌‌‍​⁠‌‍⁠⁠‌​‌⁠‌​‌​​‍​‍‌⁠⁠‌​