In short: 18 example Auto-Moderation rules, built from Zeal's AI signals and data points, listed in the order they should run.
Understanding the rule examples
These examples show how to combine conditions into Auto-Moderation rules that label posts for you, so your moderators start from a pre-sorted queue. Copy the logic, then swap in your own brand names, image features, prices and markets.
Each example shows:
Purpose: the pattern the rule is designed to catch.
Conditions: written as Condition → operator → value. A post must match all of them.
Label: the label Zeal applies when the rule matches. Labels are yours to name, so use the ones your Program already works with.
Conditions use three kinds of operator:
Any In / None In: for lists, such as keywords, image features or account labels. None In is how you exclude something, for example a post with no logo.
Is True / Is False: for yes/no signals, such as Post Abnormal Price or Post Has Obfuscated Brand Name. Choosing the operator is enough; there's nothing to tick.
Greater than / Less than: for numbers, such as Price, Post Stock Count or Account Follower Count.
Note: These rules are generic examples. Tailor the keywords, thresholds and markets to your brand and risk tolerance, and agree them with your Brand Partner before activating.
Jump to: 1. Whitelisting · 2. Impersonation and phishing · 3. Counterfeits · 4. Copycats · 5. Grey market and policy · 6. Cleaning up the queue · 7. Rule order
1. Whitelisting
Whitelisting rules stop genuine sellers being labelled as infringers, so create them first and keep them at the top of your rule order.
1.1 Your official accounts
Purpose: label posts from your own brand accounts so no other rule touches them.
Account Name → Any In → ["solene_official", "Solène Paris"]
Label: Official Store
1.2 Approved sellers
Purpose: carry over decisions your team has already made about an account to every new post from it.
Account Label → Any In → ["Authorized Seller", "Legitimate"]
Label: Legitimate
Important: If a detection rule sits above your whitelisting rules, it can label your own authorised sellers as infringers.
2. Impersonation and phishing
These rules catch accounts pretending to be your brand. They often signal fraud and can harm consumers quickly, so they run straight after whitelisting.
2.1 Fake brand accounts
Purpose: catch accounts using your brand name that you haven't approved.
Account Name → Any In → ["solene", "solène official", "solene_store", "s0lene"]
Account Label → None In → ["Official Store", "Authorized Seller", "Legitimate"]
Label: Impersonation
2.2 Fake support and refund scams
Purpose: catch accounts posing as customer support to take payments or personal details.
Account Name → Any In → ["support", "help desk", "refunds", "customer care"]
Post Description and Title → Any In → ["refund", "verify your account", "payment link", "claim your order"]
Label: Phishing
2.3 High-reach fake accounts on social media
Purpose: surface the fake accounts that can do the most damage, so they're handled first.
Account Name → Any In → ["solene", "solène official", "solene_store"]
Website Category → Any In → ["Social Media"]
Account Follower Count → Greater than → 10,000
Label: Impersonation – High Reach
Note: Website Category in Auto-Moderation rules is the same data point as the Channel filter, such as Marketplace or Social Media.
3. Counterfeits
These rules lean on Zeal's AI signals rather than long keyword lists, so they stay accurate as sellers change their wording.
3.1 AI flags an obvious fake
Purpose: act on Zeal's strongest image signal: your logo on a product the AI classes as an obvious counterfeit.
Image Features → Any In → ["Solène Logo"]
Image Feature Type → Any In → ["Obvious Counterfeit"]
Label: Counterfeit
3.2 Hidden brand name at a suspicious price
Purpose: catch sellers disguising your brand name to dodge searches while undercutting the market.
Post Has Obfuscated Brand Name → Is True
Post Abnormal Price → Is True
Label: Counterfeit
3.3 Suspicious image, suspicious price
Purpose: combine a suspicious AI image signal with price and wording signals for a high-confidence match.
Image Features → Any In → ["Solène Logo"]
Image Feature Type → Any In → ["Highly Suspicious"]
Post Abnormal Price → Is True
Post Has Obfuscated Brand Name → Is True
Label: Counterfeit
3.4 An image already confirmed as fake
Purpose: reuse your team's work: once an image has been confirmed as counterfeit, label every new post that uses it.
Image Label And Status → Any In → ["Counterfeit – Validated"]
Label: Counterfeit
3.5 Openly selling replicas
Purpose: catch sellers who name your brand and say outright that it's a copy.
Post Listed Brand → Any In → ["Solène"]
Post Description and Title → Any In → ["replica", "1:1", "mirror quality", "AAA", "super copy"]
Label: Counterfeit
3.6 Logo shown, brand left out
Purpose: catch posts that show your logo in the image but keep your brand name out of the text.
No Listed Brand → Is True
Image Features → Any In → ["Solène Logo"]
Image Feature Type → Any In → ["Highly Suspicious", "Obvious Counterfeit"]
Label: Counterfeit
3.7 Bulk new stock below retail
Purpose: catch sellers holding large volumes of "new" products at prices genuine stock can't match.
Post Listed Brand → Any In → ["Solène"]
Post Product Condition → Any In → ["New"]
Price → Less than → [your minimum genuine price]
Post Stock Count → Greater than → 50
Label: Counterfeit
4. Copycats
Copycats imitate your designs without using your logo, so these rules look for your design features and exclude anything carrying the logo itself.
4.1 Dupe marketing, no logo
Purpose: catch sellers advertising lookalikes of your products.
Post Description and Title → Any In → ["dupe", "inspired by", "lookalike", "same style as"]
Image Feature Type → None In → ["Logo"]
Label: Copycat
4.2 Your design, without your name
Purpose: catch products copying a signature design element, such as a pattern or clasp.
Image Features → Any In → ["Solène Monogram Pattern", "Solène Clasp"]
Image Feature Type → None In → ["Logo"]
No Listed Brand → Is True
Label: Copycat
5. Grey market and policy
These rules catch genuine products sold where or how they shouldn't be.
5.1 Sold in an unauthorised market
Purpose: find new stock offered by sellers in countries where you don't distribute.
Post Listed Brand → Any In → ["Solène"]
Post Product Condition → Any In → ["New"]
Account Country → Any In → [your unauthorised markets]
Account Label → None In → ["Authorized Seller", "Official Store"]
Label: Grey Market
5.2 Testers and samples
Purpose: catch the resale of products that were never meant to be sold.
Post Listed Brand → Any In → ["Solène"]
Post Description and Title → Any In → ["tester", "not for resale", "sample", "display item"]
Label: Policy Violation
6. Cleaning up the queue
These rules take low-risk posts out of your moderators' way, which is often the biggest time saving of all. Put them at the bottom so every detection rule gets a chance first.
6.1 Genuine second-hand
Purpose: set aside used items at a normal price with no counterfeit signals, which are usually lawful resale.
Post Product Condition → Any In → ["Used"]
Image Feature Type → None In → ["Obvious Counterfeit", "Highly Suspicious"]
Post Abnormal Price → Is False
Label: Second-hand
6.2 No brand signal at all
Purpose: clear posts that neither name your brand nor show your logo.
No Listed Brand → Is True
Image Feature Type → None In → ["Logo"]
Label: Not Relevant
Tip: Check a clean-up rule's Preview tab before activating it, to make sure it isn't catching posts your team would want to see.
7. Rule order
When a post matches more than one rule, the rule highest in your list applies, so the order decides the final label. Drag rules in Settings → Moderation → Auto-moderation into this order:
Order | Rules | Why here |
1 | Whitelisting: 1.1, 1.2 | Protects genuine sellers from every rule below |
2 | Impersonation and phishing: 2.2, 2.3, 2.1 | Highest consumer risk; the most specific rule first |
3 | Counterfeits: 3.1, 3.4, 3.3, 3.2, 3.5, 3.6, 3.7 | Strongest signals first |
4 | Copycats: 4.2, 4.1 | After counterfeits, so a post showing your logo is labelled Counterfeit rather than Copycat |
5 | Grey market and policy: 5.2, 5.1 | Genuine products, so only once counterfeits are ruled out |
6 | Cleaning up the queue: 6.1, 6.2 | Only catches what no detection rule matched |
Learn more: Building Auto-Moderation Rules · Rule Suggestions · AI Image Feature Detection · Channel Filter
Quick reference
If you want to… | Start with… |
Protect your own and approved accounts | 1.1 and 1.2 |
Catch fake brand accounts and scams | 2.1, 2.2 and 2.3 |
Trust Zeal's AI on obvious fakes | 3.1 |
Stop re-moderating the same fake image | 3.4 |
Catch hidden brand names and cheap fakes | 3.2, 3.6 and 3.7 |
Catch lookalike designs | 4.1 and 4.2 |
Find genuine goods in the wrong market | 5.1 |
Shrink the moderation queue | 6.1 and 6.2 |