Monitoring brand mentions is important for understanding how people talk about your brand.
If someone is happy with your brand, it’s good to jump into the conversation and amplify their post or comment.
And if someone is unhappy with your brand, you want to go in and offer support.
SocialListeningAPI searches public posts and Google results for those brand terms, then returns the source URLs in one response shape. That means you can find praise worth amplifying and complaints that need support without checking each source by hand. Your scheduler, storage, and delivery tools create the alert workflow.
Step 1: Build the brand mention query list
Create a SocialListeningAPI account and claim the 100 free credits. List:
- Brand and product names
- Domain without protocol
- Founder or public spokesperson names
- Social handles
- Common misspellings
- Campaign or feature names
Keep each query separate so the alert records what matched.
Step 2: Search brand mentions with the MCP
Connect the MCP to Claude or ChatGPT, then use a prompt like this:
Search LinkedIn, X, Reddit, Facebook, Hacker News, TikTok, YouTube, and Google for each brand term
below. Return the platform, matched term, short context, published date, and source URL. Group posts
that may need support separately from posts that may be worth amplifying. Do not draft or send a
reply.
Brand terms:
- brand name
- product name
- domain.com
- common misspellingReplace the example terms with your brand and product names.
One query across eight sources costs 15 credits. The Facebook endpoint docs explain its 7-credit search and one-page limit. LinkedIn search costs 2 credits, and each other search costs 1 credit.
Step 3: Schedule and deduplicate externally
Use cron, n8n, Make, or Zapier to run the approved queries. Store item ID or source URL, platform, query, published date, and first-seen time in an external database. Send only records not seen before.
Scheduling, storage, deduplication, and delivery happen outside SocialListeningAPI.
Facebook sentiment is disabled. Do not label a mention positive or negative without a separate external review step.
Step 4: Route the alert
Send the source URL and short context to email, Slack, a support tool, or another chosen destination. Use an external classifier when you need routing by support, PR, sales, or product topic. A human should review sensitive mentions.
Alerts, destinations, classification, and webhooks are not native SocialListeningAPI features.
If an alert leads to a Reddit reply, a human must read the community rules and approve it.
Conclusion
Begin with a manual MCP search for the brand name, domain, and one misspelling. Remove terms that return unrelated names before an external tool starts a recurring schedule.
An alert should contain the source, match reason, and first-seen time. Everything else can wait for human review.
Continue with Build a Weekly Competitor Mention Brief.
