A practical guide · ~6 min read
You can tell someone almost everything about a stretch of a person's life by what they listened to during it. The album they wore out the summer they moved. The song they queued at 2am the year they were unhappy. The artist they returned to after a breakup they told nobody about. Spotify has this in a database — years of it — and it will hand you a copy if you ask. Here's how.
Most digital exhaust is noisy — every click, every scroll, most of it forgotten by the person who made it. Music is different. People don't play a song for the fifth time in a week by accident. Repeat listens are always meaningful; they just aren't always meaningful in ways you can articulate. That's exactly the kind of signal an AI memory tool can pattern-match on where you can't.
Over a decade the resolution is startling. Life transitions leave fingerprints in the listening data: a genre suddenly appearing, an artist quietly disappearing, tempo shifting slower for a while and then coming back. If you've ever looked at a Spotify Wrapped and thought "yeah, that was the year" — that's the effect at low resolution.
Spotify offers two exports, and most people request the wrong one:
You can request both. Do request both. The extended history is the one you'll actually use.
Konshus is one way to solve this — a persistent memory vault and portable persona that follows you across ChatGPT, Claude, Gemini, and whatever ships next.
Meet KonshusJSON files listing every song stream — timestamp, track, artist, album, milliseconds played, platform, whether you skipped it. Also playlists, saved albums, follows, podcast episodes. Total size depends on how long you've had the account and how much you listen; anywhere from a few MB to several hundred.
Raw play counts aren't memory. "You played 4,721 songs by Radiohead" is trivia. What matters is the pattern: the months where they were on repeat, the season when they suddenly weren't, the year they came back. That's what a good memory tool aggregates for you.
Konshus imports Spotify history as weekly and monthly listening patterns — never raw play data — and each pattern becomes an atom your Konshus can quote back to you when you say "what was I into last summer?"
Konshus reads your Spotify history and turns years of streams into weekly patterns — memory your Konshus can reference when a song comes up.
Meet Konshus