Spam in a language you cannot read
Comments in any language are judged by what they mean, so spam you cannot read is caught the same as English.
On this page
Before you start
- The AI rules read every language; write in your own.
- Keywords are exact, so they cover only the words you add.
- Translate a hidden comment in one click to check the verdict.
Switch on the four rules that judge meaning.
Left sidebarPagesyour pageModerationthe rules with the AI mark
You will see Each rule shows examples, with a switch and an action.
Write your topics and notes in your own language.
Custom topicsyour topics, in your languageEnter
You will see Your topics as chips; the same meaning in any language is caught.
Test with a comment in a language you do not speak.
Playground, at the bottom of the tabpaste the commentrun
You will see The verdict names the rule, before anything is live.
Translate what was hidden, to check the verdict.
InboxTranslate, under the comment
You will see The comment in your language, and a verdict you can check.
If something is off
- Spam in another language got through
- Describe the spam under Custom topics; Keywords match only your words.
- A real customer wrote in their language and was hidden
- Translate, unhide, and whitelist the words that must survive.
- I want to reply but do not speak the language
- Write in your language; Translate in the composer does the rest.
- AI checks are unavailable
- The AI rules retry by themselves; the text rules keep running.
Common questions
Which languages does CommentWatchDog moderate?
Any language. The four AI rules, Profanity, Negativity, Custom topics and Impersonators, read the meaning of each comment in whatever language it arrives, including creative or disguised spellings. The text rules, URLs, Emails and phone numbers, Mentions, Hashtags, Images, Emojis and Keywords, match exactly what is in the comment, so they need no language at all.
Do I have to write my custom topics in every language my ads run in?
No. Write a topic once, in your own language. The rule judges meaning, so a comment that says the same thing in Spanish, Arabic or Hindi matches the same topic. Only the Keywords rule is a word list, and it matches only the exact words you add.
Can I read a hidden comment I do not understand?
Yes. Translate, under any comment in the inbox, turns it into your translation language, with Show original to flip back; Auto translate on the Settings tab translates every comment as it arrives. Translation starts on Starter.
Is moderation in other languages included on every plan?
Yes. The preset AI moderation, Profanity and Negativity, runs in any language on every plan, Mini and Lite included. Your own custom topics, the Negativity note and the Impersonators check start on Starter.
For agents: plain text of this guide · all guides
For agents
Guide: Spam in a language you cannot read
This page: https://commentwatchdog.com/help/moderation-rules/spam-in-a-language-you-cannot-read
Topic: Moderation rules
Time: about 3 minutes
Who can do it: Owner, Administrator or Manager
Author: https://anantgupta.xyz (Anant Gupta, founder of CommentWatchDog)
Start in the app: https://app.commentwatchdog.com/en/inbox
Result when done: Comments in any language are judged by what they mean, so spam you cannot read is caught the same as English.
Context: Ads that run in several countries collect comments in languages nobody on the team reads. The four rules that judge meaning, Profanity, Negativity, Custom topics and Impersonators, read the comment itself in whatever language it arrives, so a scam in Arabic or an insult in Portuguese is caught with no word list in that language. The rules that match text, URLs, phone numbers and Keywords, are exact and need no language at all. What was hidden can be translated in the inbox with one click, so you can check a verdict without knowing the language.
Before you start
1. The AI rules read meaning in any language; you write your topics and notes in your own language and they apply to every comment.
2. Keywords match the exact words you add, so a keyword list covers only the languages you wrote it in.
3. The inbox translates a comment into your language on demand, or automatically with Auto translate.
Steps
Step 1 of 4: Switch on the four rules that judge meaning.
Where: Left sidebar, Pages, the page, then the Moderation tab. Profanity, Negativity, Custom topics and Impersonators carry the AI mark. They read the comment and the author name, not a list, so they need no words in the commenter's language.
Link: https://app.commentwatchdog.com/en/inbox (Open the app)
Screenshot: https://commentwatchdog.com/media/shots/moderation.en.webp (The AI mark on the rules that judge meaning.)
Screenshot: https://commentwatchdog.com/media/shots/moderation-phone.en.webp (The same on a phone.)
You should see: Each rule shows examples of what it catches; the switch and the action are on the row.
Step 2 of 4: Write your topics and notes in your own language.
Where: Under Custom topics, type what you want caught in plain words, for example people selling their own tools, or anyone asking to be contacted on WhatsApp, and press Enter. Under Negativity, write the note about your business. Both are read as meaning, so a comment in another language that says the same thing matches.
Screenshot: https://commentwatchdog.com/media/shots/moderation.en.webp (Custom topics, written in your own language.)
Screenshot: https://commentwatchdog.com/media/shots/moderation-custom-phone.en.webp (The same on a phone.)
You should see: Your topics as chips. A comment in any language that means the same thing is caught.
Step 3 of 4: Test with a comment in a language you do not speak.
Where: The Playground at the bottom of the Moderation tab. Paste a real comment from under your ad, in whatever language it came, and run it. The verdict names the rule that would catch it.
Screenshot: https://commentwatchdog.com/media/shots/moderation-playground.en.webp (A comment in another language, in the Playground.)
Screenshot: https://commentwatchdog.com/media/shots/moderation-playground-phone.en.webp (The same on a phone.)
You should see: The verdict, with the rule that decided, before the rule touches a live comment.
Step 4 of 4: Translate what was hidden, to check the verdict.
Where: In the inbox, Translate under a comment turns it into your language, with Show original to flip back. Auto translate on the Settings tab does it for every comment as it arrives.
Link: https://app.commentwatchdog.com/en/inbox (Open the inbox)
Screenshot: https://commentwatchdog.com/media/shots/translate.en.webp (A translated comment, with Show original to flip back.)
Screenshot: https://commentwatchdog.com/media/shots/translate-phone.en.webp (The same on a phone.)
You should see: The comment in your language, so you can unhide a real customer or leave the spam hidden.
If something is off
Problem: Spam in another language got through
Fix: A word-list rule was doing the work. Switch on Custom topics and describe the spam as a topic; the AI rules read meaning, Keywords match only the words you added.
Problem: A real customer wrote in their language and was hidden
Fix: Translate it, unhide it, and add the words that must survive to the Whitelist. For Negativity, write the note so the rule knows what a fair complaint looks like for you.
Problem: I want to reply but do not speak the language
Fix: Write the reply in your language; Translate in the composer turns it into the commenter's language before you send.
Problem: AI checks are unavailable
Fix: The AI rules retry on their own; URLs, phone numbers and Keywords keep running meanwhile.
Common questions
Question: Which languages does CommentWatchDog moderate?
Answer: Any language. The four AI rules, Profanity, Negativity, Custom topics and Impersonators, read the meaning of each comment in whatever language it arrives, including creative or disguised spellings. The text rules, URLs, Emails and phone numbers, Mentions, Hashtags, Images, Emojis and Keywords, match exactly what is in the comment, so they need no language at all.
Question: Do I have to write my custom topics in every language my ads run in?
Answer: No. Write a topic once, in your own language. The rule judges meaning, so a comment that says the same thing in Spanish, Arabic or Hindi matches the same topic. Only the Keywords rule is a word list, and it matches only the exact words you add.
Question: Can I read a hidden comment I do not understand?
Answer: Yes. Translate, under any comment in the inbox, turns it into your translation language, with Show original to flip back; Auto translate on the Settings tab translates every comment as it arrives. Translation starts on Starter.
Question: Is moderation in other languages included on every plan?
Answer: Yes. The preset AI moderation, Profanity and Negativity, runs in any language on every plan, Mini and Lite included. Your own custom topics, the Negativity note and the Impersonators check start on Starter.
Related guides
Custom topics in plain words: https://commentwatchdog.com/help/moderation-rules/custom-topics-in-plain-words
The twelve rules, one by one: https://commentwatchdog.com/help/moderation-rules/the-twelve-rules-one-by-one
Translate a comment and your reply: https://commentwatchdog.com/help/inbox/translate-a-comment-and-your-reply
Plain text of this guide: https://commentwatchdog.com/help/moderation-rules/spam-in-a-language-you-cannot-read.md
All guides for assistants: https://commentwatchdog.com/help/llms.txt