Verification, consent, privacy controls, methodology, and delivery — exactly what AnalyzeG does, what it does not prove, and what you can verify yourself.
1. How Verification Works — and What It Does Not Prove
Every review passes 5-point screening: face-presence, device integrity, geo-consistency, content checklist, and timestamp validation plus human moderation. Checks are time-stamped and securely recorded with audit logs.
Face-presence confirms a face is present in the 30-second in-hand video; it does not process biometric templates and does not prove product ownership by itself.
Device checks for integrity signals; geo checks consistency; content checks the checklist is followed; timestamp validates recency.
Human moderation reviews flagged or borderline submissions, duplicate reviewer detection, and false-content policy enforcement.
Verification reduces risk but does not guarantee every opinion is genuine or every product claim is true. Buyers receive raw video and raw CSV to verify insights themselves. Rejection criteria include missing face, no product in hand, scripted/duplicate content, and policy violations.
2. Reviewer Consent and Video Licensing — Separate Choices
Reviewers make three separate consents at submission:
Permission to submit a review for verification.
Permission to sell research insights derived from the review.
Optional, separate permission for the relevant product's brand to licence the video for advertising. Competitors receive data insights only — never video ad rights.
Licence scope, media, duration, territory, and withdrawal are governed by our Terms. Withdrawal does not affect licences already granted; future licensing stops upon verified request to [email protected]. All licences require the reviewer has opted in.
3. Data Access and Privacy Controls
Buyers receive secure video access links with role-based access and licence controls plus raw CSV/Excel with video links, state/age/gender aggregates, sentiment and keyword themes.
Buyers do not receive reviewer phone numbers, UPI handles, addresses, or direct contact details. Demographics are provided as aggregated filters; personally identifiable details are not provided to buyers.
Access is controlled, time-bound, and logged. Data-handling is designed to support India’s data-protection requirements — DPA available on request.
4. Research Methodology
AnalyzeG recruits verified consumers who already own the product category. Eligibility is checked at submission; quotas are validated via 5-point screening + moderation.
Recruitment: opted-in community, no purchased lists. Incentivization is a flat transparent compensation per verification, not per rating or sentiment — we do not pay for positive reviews.
Bias management: structured checklist, moderation, duplicate prevention, and raw data access for independent verification.
Limitations: category and audience availability vary; we confirm feasibility before any custom collection.
What buyers receive is stated in each package: video count, Power BI template, sentiment/keyword output, and raw export. Nothing is promised beyond that.
5. Marketplace Inventory and Delivery Policy
In-stock datasets: verified reviews already available in the marketplace for listed categories. Delivery is near-instant (often minutes) after purchase and payment verification.
Custom or quota-specific requests (new product, specific state/age/gender mix, competitor comparison with custom scope): availability and timeline are confirmed before purchase. No custom collection begins until you approve scope, quota, and SLA in writing.
SLA for in-stock: delivered in 2 minutes to 24 hours. If in-stock delivery exceeds 24 hours, you receive a 10% credit. For custom, the agreed SLA in your order confirmation applies.
Refund/credit: in-stock data is licensed upon delivery and is generally non-refundable; credits apply for SLA misses or verified data defects per our Refund & Cancellation terms.
6. Sample Dataset Methodology
Every sample is labelled:
Anonymized mock — synthetic data that shows dashboard structure (e.g., the Power BI preview on the homepage at 68% Positive / 22% Neutral / 10% Negative).
Anonymized real data — when we share it, it will be explicitly marked with collection date range, category, sample size, and anonymization method.
The homepage Power BI preview and the “Illustrative Dashboard Scenario” on the homepage are mock/anonymized scenarios to demonstrate how filters reveal hidden issues — not client results.
Download the 5-row sample CSV with no email required: sample-analyzeg-reviews.csv — column definitions and video-link format are in the header.
Questions on verification, consent, or delivery? Our research lead replies within 4 hours.
