Trend Query Engine

Ask: 'Saree trends', 'Gown trends', 'Lehenga trends'. We scan supply (Tier 1), social (Tier 2), marketplace (Tier 3) and own-store (Tier 4) and return a full PEWS/DEWS report with sources, region distribution, raw materials and supplier catalogues.

Open Trend Phases · Lifecycle

Pair this with the Lifecycle view — query results are scored against the active phase (Pre-trend → Mass → Saturation) for that dress type.

Query Types — pick a template, get a full report

8 categories · 40+ templates

Each query type runs through a different signal weighting and gate. Click any chip to scan instantly.

Dress-type trend scan

Per-category PEWS with category-specific weights (saree leans on weaver supply, gown leans on Paris/Milan editorial, streetwear leans on TikTok).

PEWS = (S_supply·w₁ + S_social·w₂ + S_market·w₃ + S_own·w₄) × G_supply × A_geo

Material × design × color

Combinatorial morphological queries — Fabric × Color × Drape × Embroidery. Returns the highest-probability concept slice.

P(adoption | DNA, X) where X = (Fabric, Color, Drape, Embroidery, Neckline, Sleeve)

Region × city

A_geo amplifier locked to a region or Tier-1/2 city. Surfaces the regional adoption probability and boutique allocation.

PEWS_region = PEWS × A_geo(region) × demand_concentration(city)

Phase / lifecycle

Filter by current cycle phase: Origin, Pre-trend, Early, Mid, Mass, Saturation, Decline. Drives the GO / WATCH / HOLD / EXIT gate.

Decision = stage(PEWS, DEWS) × longevity × NOT(meesho_flood)

Channel-specific signal

Pull weighted by a single channel — useful when you want IG-led narrative or Pernia's-led editorial validation only.

PEWS_channel = Σ(signal_i × w_channel_i) for selected channel only

Inspiration / fusion

Fusion queries (Japanese × Indian, Paris × African × Indian) using NOT operator to remove saturated combos. Routes to Designer Copilot.

X_unique = (Inspiration_A ⊕ Inspiration_B) ∩ NOT(saturated ∪ marketplace_clones)

Pricing & margin

Price-band queries return raw-material BOM, supplier catalogues, and recommended MRP given the PEWS strength.

MRP_optimal = COGS × markup × (1 + PEWS_uplift) × (1 − competitor_pressure)

Exit gate / risk

DEWS-led queries — surface SKUs and trends that have crossed the kill threshold (Meesho flood, return-rate spike, editorial decay).

DEWS = decay_velocity + meesho_clone_count + return_rate − new_signal_rate