A behavioral matchmaking pilot using longitudinal Spotify data — the first real-world proof that what people do predicts compatibility better than what they say.
We're building a listening-data registry designed for the agent economy. Our core insight: AI agents need longitudinal, real-world behavioral signals to stay grounded — and music listening is one of the richest, most continuous preference streams humans generate.
Blue Sky Tech turns listening history into structured, cross-platform data that can be queried agent-to-agent (A2A). Instead of static profiles or synthetic preference guesses, agents reference an evolving behavioral record that captures taste shifts, patterns, and context over time.
The result: better agent reasoning, better personalization, and a path toward an open human data layer where people own and monetize the signals their lives produce. If it works for listening data, our underlying technology scales to any vertical where human behavior can be opted-in and captured.
"We already know — because they've been telling Spotify for years."
Every skip, replay, and late-night listening session is a revealed preference — honest in a way no survey ever is. Blue Sky Tech's technology turns that continuous behavioral record into a structured compatibility signal.
This pilot is the first real-world test of a simple but powerful claim: longitudinal behavioral data produces better human outcomes than self-reported preferences. Better matches. Better dates. Proof that what you do is more predictive than what you say.
Existing algorithm — surveys, profiles, and behavioral signals already in the app. This is the baseline.
Spotify listening history scored for taste compatibility — honest, longitudinal, zero self-reporting required.
Users describe idealized versions of themselves and partners — not real ones. Aspirational ≠ accurate.
A 5-minute quiz cannot capture years of authentic behavioral signal. People change. Quizzes don't.
Most apps measure swipes and messages. Nobody asks whether the date actually went well. We will.
Pilot participants connect their Spotify account via OAuth. We pull top artists, tracks, genres, and listening history — no personal data stored beyond what's needed for scoring. Cohort size scoped with the app team.
Multi-dimensional taste-compatibility score computed per user pair: genre graph overlap, audio feature alignment, temporal listening patterns. Music as a revealed behavioral proxy — not just "you both like the same artist."
Cohort split into control and experiment. Both groups use the app normally. After dates, both groups rate their experience. We collect, analyze, and compare. The control group sets its own bar.
Results — win or lose — become a public case study on behavioral data in matchmaking. Blue Sky Tech retains methodology IP. The portfolio company gets a validated, differentiated feature if the hypothesis holds. The industry learns either way.
Music is the world's best icebreaker. It should show up on the date itself — so we measure exactly that.
After a match goes on a date, both users are prompted to rate the experience. The control group's ratings serve as the baseline — if the experiment group rates measurably higher, the hypothesis holds.
The experiment group's average date quality score is statistically higher than the control group's at pilot close. No arbitrary number — the control group sets its own bar. If their current method is working, we'll see it.
Covers Spotify API integration, scoring engine build, 8 weeks of infrastructure, and full analysis + case study write-up. Scope-dependent — finalized after cohort sizing with app team.
Flat bonus triggered if experiment group's date quality score is statistically higher than control at pilot close. Defined and agreed before launch. Incentives aligned.
Positive results give the portfolio company right of first refusal on a full API licensing deal. Blue Sky Tech retains methodology IP. They get a proven, proprietary matching layer — we get a validated product in market.
We need to scope the integration, design the vibe-check UX, and finalize cohort size before locking pricing.
Your trust accelerates the conversation significantly — especially given the portfolio company's second-chance capital context.
Ready to move immediately. Spotify OAuth + scoring engine live within 2–3 weeks of agreement. 8-week pilot clock starts on first user authorization.
Reach out directly to either of us. We're ready to move fast.