WEBOTEE · DATA METHODOLOGY

How we measure coverage.

Webotee combines two complementary data collection regimes — one narrow and deep for your provisioned catalog, one broad and shallow for the marketplace-wide operator graph. Here is exactly what each regime does, how often it runs, and where it stops.

Per-customer targeted collector

For every ASIN in your provisioned catalog, our collector captures the Amazon product detail page and extracts the full offer listing — every seller, their price, fulfillment method, and buybox status. This pass runs at the cadence determined by your tier (see below) and is what populates the Operator and Contested-ASIN views on each Brand Dossier.

What it produces: per-ASIN seller snapshots fed into the brand catalog views powering every Brand Dossier. Coverage % reflects how many of your provisioned ASINs have been refreshed within the last 24 hours.

BSR top-1.5M operator walk

Independently of your catalog, we continuously monitor the Best-Seller Rank top 1 million ASINs across Amazon. The purpose is not per-ASIN completeness — it is cross-brand breadth. By aggregating which sellers appear across many brands and categories, we build the Operator Directory: a register of high-velocity wholesale and multi-brand sellers active across the marketplace.

What it produces: The Network View and Operator dossiers — showing which operators touch multiple brands simultaneously, even brands you don't manage. This is the primary tool for detecting systematic distribution-chain problems that pre-date authorization violations on your own catalog.

Update cadence by tier

Scout

Daily

One full catalog pass every 24 hours.

Scout Pro

Daily

One full catalog pass every 24 hours with priority-queue ordering for newly uploaded ASINs and alert-triggering events.

Scout + Protect

Daily +

Daily full-catalog pass plus accelerated refresh on specific ASIN sets on request — contact your account manager to scope.

We do not claim real-time. Webotee runs continuous daily data collection — we respect the platforms we monitor and operate within their rate limits.

History depth

Seller identity

Dec 2024 → present

Display name of the seller holding the buybox on each tracked ASIN. This window grows daily.

Price history

Up to 29 months

Nov 2023 → present. Daily price snapshot per ASIN — surfaces buybox transitions before explicit seller identity is available.

Stable seller IDs

Up to 4 months

Nov 22 2025 → present. Stable merchant IDs that survive display-name changes (important for hijacker attribution).

Your plan determines how much of this depth is exposed in the app. Contact your account manager to unlock longer windows.

Optional SP-API supplementary layer

ON REQUEST · ENTERPRISE

For Enterprise customers who enrol their own Amazon Brand Registry credentials with Webotee, we can layer Amazon's Selling Partner API feed on top of the continuous collector. SP-API data narrows the window between offer-listing changes and when those changes surface in your alert stream — useful when enforcement SLAs are measured in hours rather than days.

Status: available on request, not enabled by default. Implementation is scoped per engagement — we build only the signal types your team will actually action. Enrolment requires your authorised Brand Registry admin to grant Webotee developer-app access; Webotee never receives credentials for accounts we haven't been explicitly invited into.

What it does not do: SP-API does not replace the collection layer — it supplements it for specific high-value events (buybox-change, new-offer, listing- suppression). Most of what you see in the app remains collection- derived.

EXPLICITLY NOT COVERED

Books, Kindle, Audible, and digital media
Amazon Music, Prime Video, and streaming licenses
App Store / digital software downloads
Grocery & Fresh perishables with no stable ASIN lifecycle
Categories outside physical retail on Amazon UK / US

More marketplaces scoped on request.

SELLER SOURCING INTELLIGENCE

For partnership-driven 3P sellers.

Webotee's seller-sourcing lane reuses the same BSR top-1.5M catalog data as the brand-protection product — reframed for brand partnership evaluation rather than enforcement.

Entry-opportunity signals are computed nightly from price-softening patterns, seller-thinning trends, and catalog-maturity heuristics. A brand exhibiting simultaneous price erosion and seller exit across multiple ASINs is flagged as a potential partnership opening or distribution-gap opportunity.

Category reports aggregate brand density, buybox-day volume, and control-score distribution per Amazon root category — surfacing fragmented categories where new entrants have the highest chance of winning sustained buybox share.

Operator intelligence draws from the same cross-brand operator graph used by brand-protection customers. For sellers, this identifies which operators are stable multi-brand partners versus unauthorized high-velocity sellers — informing competitive positioning and wholesale channel strategy.

All sourcing data is drawn exclusively from our independent BSR data collection. No customer-uploaded ASINs or priority-queue data from other workspaces is included.

SOURCING SCORE

Per-ASIN sourcing composite.

The Sourcing Score is a 0–100 composite that summarizes how attractive an Amazon ASIN is for third-party sellers evaluating sourcing partnerships. It combines 24 per-ASIN signals across 5 dimensions. Each dimension is a micro-composite of 3–6 sub-signals with NULL-rescale: when a sub-signal source is unavailable, its weight redistributes to the measured sub-signals. Recomputed nightly.

Composite formula: Score = 0.20 × Velocity + 0.25 × (100 − Gating Risk) + 0.20 × (100 − Entry Friction) + 0.15 × Margin Signal + 0.10 × Brand Posture, renormalized / 0.90 (Operational Fit reserved for Pro+).

Traffic light: green ≥ 70 · yellow 40–69 · red < 40.

Velocity (20%)

Demand activity and market momentum for the ASIN.

SUB-SIGNALS

Brand pressure score (0.25) — cross-brand activity index from the operator graph

ASIN buybox density (0.20) — buybox_days / days_seen × 100

Bought past month (30d) (0.25) — log2(x+1) × 10, capped at 100

Niche BSR P50 (0.15) — 100 − log10(BSR) × 20, capped 0–100

Catalog churn (30d) (0.15) — 100 − churn percent (stable = high velocity)

Gating Risk (25%)

Barriers preventing new sellers from listing. Higher = more gated.

SUB-SIGNALS

Seller count inverse (0.30) — 100 − clamp(count × 15); fewer sellers = higher gating

Brand control score (0.25) — brand-level control composite 0–100

Top seller share (0.15) — dominant seller's buybox-day share

FBA penetration (30d) (0.15) — FBA offer share (FBA = table-stakes barrier)

Seller HHI top-3 (0.15) — Herfindahl concentration across top-3 sellers

Entry Friction (20%)

Competitive intensity at the listing level. Higher = harder to enter.

SUB-SIGNALS

Goldilocks band (0.35) — sweet-spot analysis on seller count (2–5 = low friction)

Buybox persistence (30d) (0.25) — how sticky the current winner is (higher = stickier)

Seller longevity P50 (0.20) — median seller tenure in days (entrenched incumbents)

Diversity inverse (30d) (0.20) — 100 − diversity score (low diversity = one seller dominates)

Margin Signal (15%)

Profit potential indicators. Higher = more margin opportunity.

SUB-SIGNALS (equal weight)

Inverse leakage (0.25) — 100 − buybox leakage (low leakage = MAP-protected)

Price stability (0.25) — 100 − coefficient of variation × 100

Arbitrage spread (0.25) — cross-channel price gap (Amazon vs Walmart)

Price spread inverse (0.25) — 100 − max seller-to-seller price spread (compression = bad)

Brand Posture (10%)

Brand health signals relevant to sourcing decisions.

SUB-SIGNALS

Mid-range control (0.25) — peak at 50 control score (reseller-friendly sweet spot)

Saturation slope (30d) (0.20) — seller-count trend (0 = flat = ideal)

Closeout signal inverse (0.20) — 100 − distress strength (closeout = penalized)*

Has GTIN/UPC/EAN (0.15) — legitimate product identifier present

Country-of-origin risk inv. (0.10) — no risk flag = higher score

Manufacturer-brand match (0.10) — listing manufacturer matches brand name

* Closeout signals penalize Brand Posture because our primary audience is Amazon sourcers and private-label sellers who want to avoid distressed inventory. Liquidation buyers and OA arbitrageurs should interpret a low Brand Posture with high closeout strength as an opportunity signal — the score is framing-dependent for this dimension.

NULL-RESCALE METHODOLOGY

When a sub-signal's data source is unavailable (e.g., no Walmart match for arbitrage spread, no closeout data for a brand), its weight is redistributed proportionally across the measured sub-signals in that dimension. If all sub-signals in a dimension are unavailable, the dimension scores 50 (neutral) and is marked unmeasured. The composite renormalizes across measured dimensions.

COMPOSITE SCORING

How we compute scores.

Brand Heat Score

A composite 0–100 metric summarizing a brand's marketplace competitive intensity on Amazon. Higher = hotter brand (more seller activity). Recomputed nightly during the materialized-view refresh cycle.

INPUTS + WEIGHTS

Authorized seller density (30%) — share of buybox-days held by sellers on the brand's authorized list

MAP discipline (25%) — consistency of pricing at or above the brand's MAP floors

Fragmentation (inverse) (25%) — inverse of top-5-operator buybox concentration

Operator stability (20%) — average tenure (months active) of the top 10 sellers on this brand

Score = Σ (normalized_input × weight) × 100, clamped to [0, 100]

If no authorized sellers or MAP floors are configured, those components default to a neutral midpoint (50/100) — the score still functions but is less informative. Configure authorized sellers and MAP floors for maximum accuracy.

Fragmentation Score

Measures how dispersed buybox ownership is across sellers on a brand. Higher fragmentation = more room for new entrants.

FORMULA

Fragmentation = 1 − (top-5 operators' combined buybox-day share)

Range: 0.0 (one seller holds 100%) → 1.0 (no top-5 concentration)

A fragmentation score near 1.0 means many small sellers contest the brand — potential for new authorized distribution. Near 0.0 means a small group dominates the brand's buybox, typical of well-controlled brands with tight distribution.

MAP Discipline Score

Measures how consistently a brand's Amazon pricing holds above the brand's MAP (Minimum Advertised Price) floors. Higher = stronger enforcement.

INPUTS

Below-MAP event frequency — count of (ASIN, date) pairs where buybox price < MAP floor

Authorized vs unauthorized ratio — what share of violations come from unverified sellers

Time window — full tracked period of seller-identity history (price history extends further)

Score = 100 − normalized(violation_frequency × severity_weight), clamped [0, 100]

Only computed for brands with MAP floors configured. Brands without MAP floors show "N/A" for this metric. A score of 90+ indicates tight MAP enforcement; below 50 suggests chronic below-MAP pricing from multiple sellers.

Volatility Score (Category-Level)

A 0–100 composite measuring how much churn a category experiences. Higher volatility = more instability in the category's seller and brand landscape, which can signal opportunity for new entrants.

4 INPUT COMPONENTS (equal weight, 25% each)

Seller turnover — rate of new sellers entering / exiting the category per month

Brand turnover �� rate of brands gaining or losing BSR representation

Price volatility — standard deviation of buybox prices across the category

BSR volatility — standard deviation of BSR rank movements for top-100 ASINs

Score = Σ (normalized_component × 0.25) × 100, clamped [0, 100]. Rebuilt weekly.

For brand-protection users, high volatility signals risk. For seller-sourcing users, moderate volatility (40–70) signals the best entry opportunities — enough churn to create openings, not so much that the category is unstable.

Sales Estimator

Measured vs. estimated

Amazon shows a “bought in past month” badge on roughly 17% of product listings. When the badge is present, we treat it as a measured value and pass it through directly. For the remaining ~83% of ASINs, we estimate monthly unit sales from the product’s Best Sellers Rank (BSR).

BSR-to-units curve

For each leaf category, we fit a power-law curve: units = a × BSR^b. Anchor ASINs are products that have both a BSR and a badge value. The curve is fitted via ordinary least squares on log-transformed data. Categories with fewer than 10 anchor ASINs borrow strength from the parent or root category via hierarchical shrinkage.

Rolling-median denoising

BSR can fluctuate significantly day-to-day. To avoid band-flicker (an ASIN bouncing between 1K+ and 5K+ across days), we use the 7-day rolling median of each ASIN’s BSR observations before mapping it through the category curve.

Confidence score

Each estimate carries a confidence score (0–100) reflecting the fit quality (R²), the number of anchor ASINs in the category, and how much hierarchical borrowing was needed. High-confidence estimates (>80) come from categories with many anchors and strong fits; low-confidence estimates rely more heavily on parent or seed curves.

Display threshold

Estimates below approximately 30 units per month are suppressed — Amazon’s badge does not appear at these levels, and curve-fit accuracy deteriorates in the long tail. These ASINs show no sales estimate rather than an unreliable number.

HOW FRESHNESS IS CALCULATED

The coverage bar on each Brand Dossier shows the time elapsed since the last successful data refresh for that brand's marketplace. Freshness is coloured green when the last refresh is under 24 hours old, amber between 24–48 hours, and red beyond 48 hours. A brand that shows 0 provisioned ASINs has not yet been tracked — contact your account manager to confirm provisioning.