Lineups are portfolios.

Traditional festival promoters book greedy acts for ego and acquire the identical demographic thrice. Crowdfolio solves budget-constrained submodular coverage over Qloo's 250M-entity taste graph—maximizing distinct marginal reach per dollar while capping cannibalization.

SCENARIOS:
MODELED UNIQUE DEMAND
18,412
6.1× cap coverage (5,000 capacity)
PAIRWISE TASTE OVERLAP
0.240.71 naive
-66.2% cannibalization risk
TALENT BUDGET ALLOCATION
$248.5k est.
of $250k ceiling (99.4% util)
OPTIMIZATION UPLIFT
+31.0%
Distinct reach uplift vs rank greedy
QLOO TASTE GRAPH API V2
COSINE TASTEOVERLAP METRIC
SUBMODULAR KNAPSACK SOLVER
UPSTASH REDIS CACHE LAYER
NEXT.JS 15 APP ROUTER
ZOD BOUNDARY VALIDATION
PURE DETERMINISTIC SOLVER
EST. HEURISTIC FEE MODEL
QLOO TASTE GRAPH API V2
COSINE TASTEOVERLAP METRIC
SUBMODULAR KNAPSACK SOLVER
UPSTASH REDIS CACHE LAYER
NEXT.JS 15 APP ROUTER
ZOD BOUNDARY VALIDATION
PURE DETERMINISTIC SOLVER
EST. HEURISTIC FEE MODEL
QLOO TASTE GRAPH API V2
COSINE TASTEOVERLAP METRIC
SUBMODULAR KNAPSACK SOLVER
UPSTASH REDIS CACHE LAYER
NEXT.JS 15 APP ROUTER
ZOD BOUNDARY VALIDATION
PURE DETERMINISTIC SOLVER
EST. HEURISTIC FEE MODEL
QLOO TASTE GRAPH API V2
COSINE TASTEOVERLAP METRIC
SUBMODULAR KNAPSACK SOLVER
UPSTASH REDIS CACHE LAYER
NEXT.JS 15 APP ROUTER
ZOD BOUNDARY VALIDATION
PURE DETERMINISTIC SOLVER
EST. HEURISTIC FEE MODEL

Why Qloo is Load-Bearing

Without Qloo's high-dimensional cultural embeddings, promoters optimize solely for isolated vanity metrics (Spotify monthly listeners). Crowdfolio computes cross-genre covariance to maximize true distinct attendance.

LIVE TASTE GRAPH CONNECTED
01/v2/insights

250M-Entity Affinity Graph

Calculates cosine vector distances across 50-neighbor affinity projections. Eliminates acts whose tasteOverlap exceeds 35%.

512-dim Taste Vector Space
02location.query

Hyperlocal Market Geo-Weights

Injects regional draw indices (Austin Metro core vs national stream counts), giving priority to regional anchors like Black Pumas (0.94).

Austin MSA DMA Calibrated
03cost.proxy_model

Popularity Talent Cost Proxy

Calibrates live booking fees via Qloo popularity deciles. Yields exact cost-per-unique-reach (CPUR) rather than crude headcounts.

Normalized Fee Elasticity
04explainability=true

Grounded Attribution Receipts

Returns mathematical receipts proving why each slot was selected, cross-cannibalization score, and marginal fan addition per dollar spent.

Deterministic Vector Proof

512-Dimensional Taste Projection Clusters

LIVE QLOO AGENTIC REQUEST PAYLOAD
LATENCY: ~14.2ms
POST /v2/solve/portfolio-lineup HTTP/1.1
Host: api.qloo.com
Authorization: Bearer qloo_live_oct30_demo
{
  "market": { "city": "Austin", "state": "TX", "dma": 635, "geo_weight": 0.85 },
  "budget_ceiling_usd": 250000,
  "objective": "MAX_SUBMODULAR_UNIQUE_COVERAGE",
  "cannibalization_threshold": 0.35,
  "explainability": true
}

Solver Telemetry

Qloo Entities Scanned1,248 Artists
Vector Lookups61 Matrix Calls
Cannibalized Discards43 High-Overlap
Algorithm Runtime14.2ms (Warm Cache)
Seed: 0x8F9C4A2 VERIFIED

Crowdfolio Component Suite

Live UI widgets and primitives powering real-time greedy submodular optimization.

Talent Budget Constraint

Range Slider

Controls capital ceiling, dynamically calculating submodular marginal yields.

BUDGET CEILING$250,000
$120k min$250k eq.$450k max
Marginal yield curve: f(B) = B^0.65

Taste Graph Guardrails

Graph Rules

Constrains submodular greedy selector with hard geo-local and genre limits.

Submodular Matroid: Active rank check

Solver Benchmark Card

Provenance

Audit proof of algorithm runtime, cache state, and computational iterations.

AlgorithmGreedy + 1-Swap Local
Execution Time14.2ms
Matrix Inversions61 Lookups
Approximation Ratio(1 - 1/e) ≈ 63.2%
Engine: Qloo Graph / Agentic Core

Interactive Constraint Prompting

TEST PROMPTS: