Sanmiraa
SANMIRAA
Log InGet Started
HomeLegal & PoliciesContent Moderation Policy
Official Document

Content Moderation Policy

Last updated: June 2026 · Sanmiraa Technologies Pvt. Ltd.

SANMIRAA

Technologies Pvt. Ltd.

MODERATION POLICY

ПОЛИТИКА МОДЕРАЦИИ

Portfolio & Content Review Rules / Правила проверки портфолио и контента

INTERNAL DOCUMENT — ВНУТРЕННИЙ ДОКУМЕНТ

Last updated / Последнее обновление: June 2026

sanmiraa.com | sanmiraa2026@gmail.com

ENGLISH VERSION

This Moderation Policy (“Policy”) sets out the internal standards, procedures, and decision-making criteria used by Sanmiraa Technologies Pvt. Ltd. (“Sanmiraa”) moderation staff and automated systems to review Designer and Supplier portfolios, Platform Listings, user-generated content, and reported material on the Sanmiraa Platform (sanmiraa.com).

This document is intended for internal use by Sanmiraa’s Trust & Safety, Moderation, and Quality teams, as well as for reference by third-party moderation contractors operating under a confidentiality agreement with Sanmiraa. Selected principles from this Policy are reflected in the public-facing Acceptable Use Policy and Terms of Use, but this document contains operational detail not published externally.

1. Purpose and Scope

Moderation exists to protect the quality, safety, and trustworthiness of the Sanmiraa Platform for all users. This Policy governs moderation of:

Designer and visualizer registration applications and Portfolio submissions.

Supplier registration applications and product Listings.

Ongoing Portfolio updates and new project uploads by existing Designers.

User-generated content including reviews, messages, and reported communications.

AI-generated content shared publicly on Profiles or in marketing-eligible galleries.

Content and accounts flagged through the in-Platform reporting system or external complaints.

2. Moderation Team Structure

2.1 Tiers of Review

AI-assisted filters perform an initial pass on all submitted content, flagging items that match known patterns of policy violation, low image quality, plagiarism signatures, or prohibited content categories.Tier 1 – Automated Screening:

Trained moderation staff review all flagged content, all new Designer/Supplier applications, and a random quality-assurance sample of approved content (minimum 10% of auto-approved submissions per week).Tier 2 – Human Moderators:

Complex, high-risk, or appealed cases are escalated to senior moderation staff or the Trust & Safety lead for final determination.Tier 3 – Senior Review:

2.2 Moderator Independence

Moderators must make decisions based solely on the criteria set out in this Policy. Moderators must not be influenced by commercial considerations, including a Designer’s subscription tier, revenue contribution, or relationship with Sanmiraa staff. Any moderator who has a personal or financial relationship with a user under review must recuse themselves and escalate to a colleague.

3. Designer and Visualizer Onboarding Review

3.1 Identity Verification

Confirm submitted identity documents are valid, unexpired, and match the name and details provided in registration.

Cross-check submitted documents against known patterns of fraudulent document submission (mismatched fonts, inconsistent metadata, template reuse across multiple applications).

Verify that the applicant has not previously been removed from the Platform under a different account (cross-reference against the internal ban list using available identity signals).

3.2 Portfolio Quality Review

Moderators assess submitted Portfolio work against the following criteria:

evidence of genuine design skill, spatial reasoning, and material/colour coordination consistent with professional-level work.Technical competence:

reverse image search and visual similarity matching are run against each submitted portfolio image to detect potential plagiarism from public design databases, competitor platforms, and stock image libraries.Originality indicators:

the portfolio should reflect a coherent body of work; a portfolio combining drastically inconsistent skill levels or styles inconsistent with claimed experience triggers manual review.Consistency:

images must meet minimum resolution thresholds (1200px minimum on the shortest edge) and must not contain visible watermarks from third-party platforms.Resolution and presentation standards:

3.3 Approval Outcomes

Outcome

Criteria

Action

Approved

Identity verified; portfolio meets quality and originality standards

Profile activated; designer notified within 48 hours

Conditionally approved

Minor issues (e.g. low-resolution images, incomplete profile fields)

Designer asked to correct specific items within 7 days before activation

Rejected – quality

Portfolio does not demonstrate sufficient professional competence

Application declined; reapplication permitted after 90 days with updated portfolio

Rejected – fraud risk

Plagiarism detected, falsified credentials, or identity mismatch

Application declined; account flagged; no reapplication without senior review approval

Standard review SLA: initial decision within 5 business days of complete application submission.

4. Ongoing Portfolio and Project Upload Moderation

4.1 Sampling Approach

Not every new project upload by an established Designer is manually reviewed. Sanmiraa applies a risk-based sampling model:

New Designers (first 90 days, or first 5 projects): 100% manual review of uploads.

Established Designers in good standing: automated screening plus a 10% random manual sample.

Designers with prior moderation flags in the last 6 months: 100% manual review until the flag-free period requirement is met (minimum 60 days).

4.2 Automated Screening Triggers

The following automatically route content to Tier 2 human review regardless of Designer history:

Visual similarity match above 85% confidence against an existing image in Sanmiraa’s reference database or public web index.

Presence of third-party logos, watermarks, or branding within the uploaded image.

User reports received on the specific upload.

Content flagged by Sanmiraa’s AI Usage Policy classifier as undisclosed AI-generated content presented as original photography or hand-drawn work.

5. Supplier and Product Listing Moderation

5.1 New Supplier Vetting

Verify business registration documentation against the jurisdiction’s relevant company registry where electronic verification is available.

Cross-check the supplier’s submitted product catalogue sample against Sanmiraa’s prohibited products list (see Acceptable Use Policy Section 4 and Supplier Agreement Section 4).

Confirm the supplier has provided valid banking/payment details for escrow disbursement purposes.

5.2 Listing-Level Review

Each new product Listing is screened for:

Prohibited product categories (weapons, counterfeit goods, hazardous materials, sanctioned items).

Accuracy red flags: pricing significantly below typical market rate for the stated product/brand combination (potential counterfeit indicator).

Image authenticity: stock images used without disclosure, or images bearing competitor marketplace watermarks.

Required compliance documentation for regulated categories (e.g. furniture flammability certification, electrical safety marks on lighting products).

5.3 Periodic Re-Verification

Active Supplier accounts undergo a re-verification check every 12 months, including a refresh of business registration status and a review of the Supplier’s cumulative complaint and return-rate history.

6. Content Moderation Standards

6.1 Prohibited Content Categories (Internal Reference)

Moderators apply the following category definitions when reviewing flagged content, consistent with the public Acceptable Use Policy:

content depicting or facilitating illegal activity. Zero tolerance; immediate removal and escalation to senior review for potential law enforcement referral.Category A – Illegal content:

reproduction of copyrighted material, branded products, or trademarked designs without authorisation. Removal with strike recorded against the account.Category B – IP infringement:

hate speech, harassment, discriminatory material. Zero tolerance; immediate removal and account review.Category C – Harmful/discriminatory content:

false claims, manipulated images, fake credentials. Removal with strike; repeat occurrence escalates to suspension review.Category D – Misleading content:

low-resolution images, incomplete listings, formatting errors. Content returned to user for correction; no strike for first occurrence.Category E – Quality/presentation issues:

6.2 Review and Rating Moderation

Moderators reviewing flagged reviews assess:

Whether the review describes a genuine, verifiable transaction (cross-referenced against order records).

Whether the review content violates content standards (harassment, defamation, irrelevant content, personal information disclosure).

Whether there is evidence of review manipulation (coordinated posting patterns, reviewer account creation immediately before posting, language indicating incentivisation).

Genuine negative reviews reflecting a real but unfavourable experience must not be removed solely because the subject of the review requests it. Removal is appropriate only where the review violates content standards or is demonstrated to be inauthentic.

7. Strike System and Escalation

Sanmiraa applies a progressive strike system for content and conduct violations:

Strike Count

Severity

Action

1st strike

Low–Medium

Warning issued; content removed; no further account action

2nd strike (within 12 months)

Medium

Formal warning; 7-day feature restriction (e.g. no new Listings)

3rd strike (within 12 months)

High

14-day account suspension; mandatory review call with Account Standards team

4th strike (within 12 months)

High–Critical

Permanent account termination, subject to senior review

Any single Category A or zero-tolerance violation

Critical

Immediate permanent termination regardless of strike history

Strikes expire from the active count after 12 months of compliant behaviour but remain in the permanent account history for reference in future moderation decisions.

8. Appeals Process (Internal Procedure)

Users may appeal a moderation decision within 14 days of notification. The internal appeals procedure is:

Appeal received via sanmiraa2026@gmail.com or the in-Platform appeal form is logged in the moderation case management system within 1 business day.

The appeal is assigned to a moderator who was not involved in the original decision.

The reviewing moderator re-examines all original evidence plus any new evidence submitted by the user.

A decision is issued within 5 business days: uphold original decision, overturn original decision, or modify the action taken (e.g. reduce a suspension to a warning).

If the user remains dissatisfied, the case may be escalated once to the Trust & Safety lead for a final internal review, which is binding within the Platform’s internal process.

9. Quality Assurance and Moderator Performance

All Tier 1 automated decisions are subject to a minimum 10% weekly human audit sample to monitor false positive and false negative rates.

Tier 2 moderator decisions are subject to inter-rater consistency checks: a minimum of 5% of decisions per moderator per month are independently re-reviewed by a second moderator.

Significant disagreement rates (above 15% on audited samples) trigger additional training and a temporary increase in senior oversight for the affected moderator.

Moderation guidelines are reviewed and updated quarterly based on emerging content patterns, new fraud techniques, and outcomes of the appeals process.

10. Confidentiality and Data Handling

Moderation staff and contractors must:

Access user data and content only to the extent necessary to perform moderation duties.

Not discuss specific moderation cases outside of authorised internal channels.

Not share screenshots, exports, or details of flagged content outside Sanmiraa’s internal systems, except as required for law enforcement cooperation authorised by senior management.

Retain moderation case records in accordance with Sanmiraa’s data retention schedule and Privacy Policy.

11. Relationship to Public-Facing Policies

This Policy operationalises the principles set out in Sanmiraa’s public Acceptable Use Policy, Terms of Use, AI Usage Policy, and Designer & Freelancer Agreement. In the event of any inconsistency between this internal Policy and a public-facing document, the public-facing document governs the rights and obligations of users, and this Policy is updated to maintain consistency.

12. Amendments

This Policy is reviewed and updated quarterly by the Trust & Safety team, or more frequently in response to emerging risks. Material changes affecting user rights are reflected in the corresponding public-facing policy updates in accordance with their respective amendment provisions.

13. Contact

For questions regarding this internal Policy, escalations, or moderation-related enquiries:

Email: sanmiraa2026@gmail.com

Website: sanmiraa.com

Legal entity: Sanmiraa Technologies Pvt. Ltd.