Canonical URL: https://chargezen.com/docs/spark/ai-agents/listening-agent # Listening Agent AI-powered 24/7 brand mention monitoring across all social platforms Last updated: 2026-01-19 The Listening Agent monitors the web 24/7 for mentions of your brand, products, and campaigns. Using visual AI and text analysis, it captures both tagged and untagged mentions—including your products appearing in images and videos where you're never tagged. ##### 142+ Untagged Mentions Monthly On average, the Listening Agent detects 142+ untagged brand mentions per month that you'd otherwise miss. These are organic conversations about your brand that traditional monitoring tools can't find. ## What It Does The Listening Agent operates continuously across all major platforms: * **Platform Scanning** — Monitors all major social platforms every 15 minutes * **Text Analysis** — Detects brand mentions in captions, comments, and posts * **Visual AI** — Identifies your products in images and videos * **Audio Transcription** — Transcribes video/audio for spoken brand mentions * **Sentiment Analysis** — Classifies mentions as positive, negative, or neutral * **Affiliate Attribution** — Links mentions to existing affiliates * **Alert Generation** — Notifies you of important mentions in real-time ## Monitored Platforms The Listening Agent covers all major platforms where your brand might be mentioned: ### Social Media * **Instagram** — Posts, Reels, Stories, comments, captions * **TikTok** — Videos, comments, sounds, hashtags * **YouTube** — Videos, Shorts, comments, descriptions * **Twitter/X** — Tweets, replies, quote tweets, threads * **Facebook** — Posts, comments, groups, Reels * **Pinterest** — Pins, boards, comments * **LinkedIn** — Posts, articles, comments (business mentions) ### Web & Blogs * **WordPress Blogs** — Articles, reviews, mentions * **Substack** — Newsletter content * **Medium** — Articles and publications * **Personal Websites** — Blog posts, reviews * **Review Sites** — Product reviews and ratings ### Other Sources * **Podcasts** — Transcribed for brand mentions * **Reddit** — Subreddits, comments, threads * **Forums** — Industry-specific discussion boards * **News Sites** — Press mentions and articles ## Detection Types The agent detects multiple types of brand mentions: ### Tagged Mentions * **@Mentions** — Direct account tags in posts or stories * **Hashtags** — Brand hashtags and campaign hashtags * **Tagged Posts** — Photos where your account is tagged ### Text Mentions * **Brand Name** — Your brand name in text (without @ tag) * **Product Names** — Specific product mentions * **Misspellings** — Common misspellings of your brand * **Abbreviations** — Shortened versions of your brand name ### Visual Mentions (AI-Powered) * **Product Recognition** — Your products appearing in photos * **Logo Detection** — Your logo visible in images * **Packaging Recognition** — Your product packaging in content * **Video Product Placement** — Products appearing in video content ##### Visual AI Training For best visual detection, upload high-quality images of your products, logos, and packaging to **Listening Agent → Settings → Visual Training**. The AI uses these as reference to find your products in creator content. ### Audio/Video Mentions * **Spoken Mentions** — Brand name spoken in videos * **Podcast References** — Mentions in podcast episodes * **Review Videos** — Unboxing or review content ## Sentiment Analysis Every detected mention is analyzed for sentiment: ### Positive Sentiment * **Indicators:** Praise, recommendations, enthusiasm, love * **Examples:** "Obsessed with this product!", "Best purchase ever" * **Action:** Flag for UGC collection, potential affiliate recruitment ### Negative Sentiment * **Indicators:** Complaints, disappointment, criticism * **Examples:** "Disappointed with quality", "Wouldn't recommend" * **Action:** Priority alert for customer service response ### Neutral Sentiment * **Indicators:** Factual mentions, questions, comparisons * **Examples:** "Has anyone tried \[brand\]?", "Comparing \[brand\] vs \[competitor\]" * **Action:** Monitor, potential engagement opportunity ## Alert Configuration Configure which mentions trigger notifications. Access at **Listening Agent → Settings → Alerts**. ### Alert Types * **High-Follower Mentions** — When creators above a threshold mention you * **Negative Sentiment** — Any negative mention for quick response * **Viral Content** — When mentions exceed engagement thresholds * **Competitor Mentions** — When you're mentioned alongside competitors * **New Untagged Mention** — First-time visual/text detection * **Affiliate Content** — When existing affiliates post about you ### Configurable Thresholds * **Follower Count** — Alert when mention from 10K+, 50K+, 100K+ followers * **Engagement** — Alert when post gets 1K+, 10K+, 100K+ engagements * **Sentiment Score** — Alert on sentiment below threshold * **Velocity** — Alert on multiple mentions in short timeframe ### Alert Delivery * **Real-time** — Instant notification for high-priority mentions * **Daily Digest** — Summary email of all mentions * **Weekly Report** — Comprehensive weekly analysis * **Channels:** Email, Slack, SMS, in-app notifications ## Automated Actions Configure automatic responses to detected mentions: ### Save to UGC Library * **Trigger:** Positive mention with high-quality content * **Action:** Automatically save to your UGC library * **Options:** Auto-tag by product, campaign, sentiment ### Tag Affiliate * **Trigger:** Mention detected from known affiliate * **Action:** Auto-attribute content to their profile * **Benefit:** Track affiliate content production automatically ### Send Thank You * **Trigger:** Positive mention from non-affiliate * **Action:** Auto-comment or DM with appreciation * **Settings:** Message templates, frequency limits, platform rules ### Request Content Rights * **Trigger:** High-quality UGC detected * **Action:** Auto-send rights request message * **Tracking:** Manages pending, approved, declined requests ### Invite to Affiliate Program * **Trigger:** Positive mention from high-fit non-affiliate * **Action:** Send program invitation or queue for Discovery Agent * **Logic:** Respects fit score thresholds from Discovery Agent ## Competitor Monitoring Track mentions of competitors to understand market positioning: * **Setup:** Add competitor brand names in settings * **Tracking:** Volume, sentiment, creator overlap * **Comparison:** Share of voice vs. competitors * **Opportunities:** Identify creators mentioning competitors but not you ##### Competitive Intelligence Use competitor monitoring to identify creators who promote competitors but might be open to your brand. The Discovery Agent can automatically evaluate these creators and suggest recruitment. ## Configuration Settings Access settings at **AI → Agent Insights → Listening Agent → Settings**. ### Monitored Keywords * **Brand Names** — Your brand and variations * **Product Names** — Individual product names * **Campaign Hashtags** — Active campaign hashtags * **Competitor Names** — Competitor brand names to track * **Negative Keywords** — Exclude irrelevant matches ### Visual Detection Training * Upload product images (multiple angles recommended) * Upload logo variations * Upload packaging images * Mark distinctive visual features ### Scanning Frequency * **Real-time** — Every 15 minutes (default) * **Hourly** — Once per hour (lower resource usage) * **Daily** — Once per day (basic monitoring) ## Performance Metrics Track listening effectiveness in the Analytics dashboard: #### Key Metrics * Total Mentions — All mentions detected (by period) * Tagged vs. Untagged — Breakdown of detection types * Sentiment Distribution — % positive, negative, neutral * Platform Breakdown — Mentions by platform * Reach — Total potential impressions from mentions * Top Mentioners — Creators mentioning you most * UGC Collected — Content saved to library * Response Rate — % of mentions you engaged with ## UGC Library Integration Detected content flows into your UGC library for organization and use: * **Auto-tagging** — Content tagged by product, sentiment, platform * **Rights Tracking** — Manage content usage permissions * **Favorites** — Mark high-quality content for campaigns * **Export** — Download content for marketing use ## Best Practices ### Train Visual Detection Well * Upload 10+ images per product from different angles * Include lifestyle shots, not just product-only photos * Update training data when launching new products ### Respond to Negative Mentions Quickly * Set up real-time alerts for negative sentiment * Have response templates ready for common issues * Escalate to customer service when needed ### Engage Positive Mentioners * Thank organic promoters promptly * Request content rights while enthusiasm is fresh * Consider affiliate program invitation for frequent mentioners ## Troubleshooting ### Missing Mentions * Check keyword configuration for typos * Add common misspellings of your brand * Verify platforms are enabled in settings * Check if scanning frequency is set correctly ### Too Many False Positives * Add negative keywords to filter irrelevant matches * Improve visual training data quality * Adjust sentiment thresholds ## Next Steps [ ### Social Listening Configure listening settings Learn more ](https://chargezen.com/docs/spark/social-listening/overview)[ ### UGC Collection Organize detected content Learn more ](https://chargezen.com/docs/spark/social-listening/ugc-collection) ### Related Articles [ #### Social Listening Listening features overview ](https://chargezen.com/docs/spark/social-listening/overview)[ #### UGC Collection Collect creator content ](https://chargezen.com/docs/spark/social-listening/ugc-collection)[ #### AI Agents Overview All agents ](https://chargezen.com/docs/spark/ai-agents/overview) Was this page helpful? Need more help? [Contact support](https://chargezen.com/docs/checkoutos/troubleshooting/support)