wGrow
menu
Crew 05 Live · Product selection + sourcing intelligence

Six analysts brief the buy. The buyer reads, signs, and orders.

What to stock. Where it sells. Which price band has air left. When the seasonal spike actually lands. Who already owns the listing. Why their reviews complain. The Selection Crew is six narrow research analysts running the brief stack a sourcing or category buyer needs before a single PO goes out — across any market the buyer asks for, in English ⇌ 简体中文 end-to-end.

We packed the seats of a category-research team into one agent crew, and gave the buyer the only seat that signs. This is one of six sample crews we run today. We'll build the seventh for whatever your sourcing function spends a week each month doing by hand.

01 The shape

A queue of briefs. Six analysts behind it. A buyer in front of it.

brief queue

The buyer asks a question — "should we list bluetooth speakers in Indonesia?", "how does business backpack demand swing in Malaysia?" — and the right analyst picks it up. Jobs run async. The console auto-refreshes. Nothing is hand-cranked.

six analysts

Market intelligence, trend forecast, competitor radar, keyword research, arbitrage scout, sentiment analyst. One narrow brief per agent. None of them tries to be a strategist.

human buyer

Reads the brief. Downloads the .docx. Signs the buy. The crew does the legwork; the human owns the procurement decision and the inventory risk that follows it.

Research Jobs queue: 16 jobs across Trend Forecast and Product Research, filterable by Running, Completed, Failed, Cancelled; auto-refreshing; bilingual EN/中文 toggle.
fig. 01 · brief queue async · auto-refreshing · bilingual
02 The agent crew

Six narrow analysts. One human buyer.

  1. market.analyst
    Market intelligence

    Target market plus product. Returns competitive market share, pricing bands by tier (entry / mid / premium), online-vs-offline channel split, executive summary. Output is a dated, downloadable brief — copyable, not a screenshot.

  2. trend.analyst
    Trend forecast

    Target market, category, forecast horizon. Returns a stable / rising / cooling verdict, price bands by tier, and seasonal demand multipliers tied to the local calendar — 6.6 and 7.7 mid-year, Merdeka Day, school-holiday windows. Multipliers are numbers (3×, 4×, 4.5×), not adjectives.

  3. competitor.radar
    Competitor radar

    Named competitors, named monitoring focus. Tracks listing velocity, price moves, assortment changes, claim shifts on the brands the buyer cares about. The agent runs against brands the buyer wrote down — not a "category landscape."

  4. keyword.research
    Keyword research

    The destination market in its own language. Surfaces the search terms buyers actually type, with intent shape (browse vs buy vs compare) and seasonal lift. Listing copy lands on the search, not next to it.

  5. arbitrage.scout
    Arbitrage scout

    Base market and unit cost in. One or many target markets out, with the realistic landed-price spread, target tier the unit lands in, and the margin band that survives shipping, duty and platform fees. Sourcing intelligence the buyer can defend in a P&L review.

  6. sentiment.analyst
    Sentiment analyst

    Mines reviews on the competitor brands the buyer named, surfaces the recurring praise and the recurring complaints. The next supplier conversation already knows what to ask the factory to fix.

  7. category buyer (human)
    Reads, signs, orders

    The brief is downloaded, read, and signed off by a human before procurement moves. The crew does not raise POs, sign MOQs, or commit working capital. It briefs; the buyer commits.

03 What it ships

Each brief is dated, sourced, downloadable. In the buyer's language.

market intelligence brief

Competitive share, pricing bands, sales-channel split, plus an executive summary that names the structural drivers — not "growth is robust." Cited back to listings and channels.

trend forecast brief

Price-band-by-tier with currency-anchored ranges, seasonal-demand multipliers tied to local anchors, competitive market-share split. The forecast horizon is on the brief, not implied.

arbitrage + sourcing brief

Base-market unit cost mapped into the target-market price spread, with the surviving margin band after fees. Where the sourcing wedge actually is, not where the marketing deck says it is.

Market intelligence brief: bluetooth speakers in Indonesia. Market-share pie (Sony, JBL, Xiaomi, Edifier, Marshall, Olike, Robot, plus locals); pricing bands USD entry / mid / premium with base and spread; sales channel split online vs offline; executive summary.
fig. 02 · market intelligence brief indonesia · bluetooth speakers · english
Trend forecast brief: business backpack in Malaysia, stable trend. Price bands $100–$200 budget tech brands, $280–$480 mid accessible luxury, $800–$2500 premium big brands. Seasonal demand: 6.6 mid-year 4.5×, 7.7 mid-year 4×, June school holidays 2.5×, Merdeka Day 3×. Competitive market share split.
fig. 03 · trend forecast brief malaysia · business backpack · 3-month horizon
04 How a buyer turns the briefs into a buy
  1. Pick what to stock. The market brief shows online vs offline split, pricing-band depth, and which tiers are saturated. The buyer reads the saturation map and lands on the tier that still has air.
  2. Time the buy. Trend forecasts tie to local seasonal anchors — 6.6 and 7.7 mid-year, Merdeka Day, June school holidays. The multiplier is a number on the brief, so order quantity and order timing both have a defensible figure.
  3. Set the price. Arbitrage scout maps a base-market unit cost into the target-market spread and shows where the margin band sits after fees. The buyer prices into the band, not against the air.
  4. Beat the listing. Keywords research surfaces the language buyers in the destination market actually use, in their own language, so the listing copy matches the search instead of fighting it.
  5. Read the room. Sentiment analyst pulls reviews on the named competitors and surfaces the recurring complaints. The next supplier brief tells the factory exactly which complaint your unit is supposed to fix.
05 What we won't ship
  1. No fabricated market shares or price bands. Every chart links back to the listings, channels and signals it pulled from. If a source rots, the number is flagged on the next refresh, not silently kept.
  2. No supplier roster pretending to be an agent. The crew briefs the buy. Supplier conversations, MOQs, sample policy, contract terms are the human buyer's job. We won't simulate a relationship that has to be earned.
  3. No "trending" without a horizon. Every forecast names its forecast horizon and the seasonal anchors it weights. "It's hot right now" is not a brief.
  4. No competitor radar without named competitors. The agent runs against brands the buyer wrote down, with a stated monitoring focus. We won't paint a "category landscape" that is really a guess.
  5. No buy decision without the buyer signing. Every brief is downloaded, read, and signed off before procurement moves. The crew sources intelligence; the human commits inventory.
06 Reusable pattern

Cut your bad-bet rate, not your category coverage.

This is one of six sample crews we run today. The other five run weekly BD pipelines, draft and fact-check articles, operate biological water treatment, ship listing-ready ecommerce imagery, and reconcile finance ops. Same shape every time: narrow agents, a curated source registry, an eval harness, a human in the only seat that signs. We'll build the seventh for whatever your sourcing function is currently doing by hand.

Brief us →