Most of what people say about your brand happens on platforms where you have no notification — Reddit threads, podcast episodes, YouTube reviews, niche trade publications. The Brand Mention Monitor aggregates these unlinked mentions across the open web, classifies sentiment, alerts on shifts, and surfaces opportunities — reviews to amplify, complaints to address, mentions to convert into backlinks. This guide covers the data sources, the classification model, the response workflow and the reclamation playbook.
The monitor continuously crawls and queries known mention surfaces — news APIs, Reddit, Twitter/X, LinkedIn, Medium, YouTube descriptions and comments, podcast show notes, trade publication RSS, plus open web mentions discovered via Bing News and brand-search crawls. Every mention is captured with its source URL, author, date, content excerpt, and a sentiment classification. The output is a stream of unlinked-but-mentioned events that you can act on.
Sentiment is graded on a five-point scale: very negative, negative, neutral, positive, very positive, with a confidence score per classification. Confidence below 0.7 means the model wasn't sure — surface for human review. Mixed-tone mentions (e.g. "Their product is great but their support is awful") are flagged as Mixed and decomposed into per-aspect sentiment where possible.
Not every mention needs a same-day response. Triage by severity and reach:
| Tier | Definition | SLA |
|---|---|---|
| Crisis | High severity, high reach (viral negative, legal threat, safety claim) | Acknowledge 2-4h, full response 24h |
| High | High severity OR high reach | 24-48h |
| Standard | Substantive but contained mentions | 3-5 business days |
| Background | Drive-by social, low-reach generic complaints | 5-10 business days OR no response |
| Positive | Capture for amplification | 3-5 days thank-you; 2-4 weeks for case-study expansion |
Every unlinked mention is a potential backlink. The Monitor surfaces unlinked mentions sorted by source domain authority — your reclamation outreach prioritises the highest-DA prospects. Typical conversion rates: 15-30% with personalised outreach, 3-8% with template outreach. See the dedicated link reclamation fix guide for the full workflow.
For each competitor you track, the Monitor counts mentions per platform, per topic, per audience segment. Where competitor mention counts substantially exceed yours, that's a coverage gap. Categorise gaps by type — platform (you're absent from Reddit), topic (you don't appear in conversations about X), audience (you have no presence with enterprise IT) — because each gap type has a different remediation playbook.
The monitor alerts when sentiment ratio deviates from baseline by more than your configured threshold. Typical thresholds: 20% drop in positive ratio over 2 weeks, or 50% rise in negative volume over 2 weeks. Diagnose root cause — incident, organic decay, viral negative, coordinated campaign, or measurement artefact — before deciding response tier.
Sources (RSS, APIs, crawls)
│
▼
Deduplication & language detection
│
▼
Sentiment classifier + confidence scoring
│
▼
Reach & severity scoring
│
▼
Triage queue (Crisis / High / Standard / Background / Positive)
│
▼
Alerts (SMS / Slack / email per tier)
│
▼
Action workflow (response templates, link reclamation, amplification)
Open-web monitoring is rate-limited and API-priced. Plan capacity:
Total monthly cost for a brand with ~500 mentions/month: typically $30-$150 in tooling.
Aggregate unlinked mentions, classify sentiment, surface response and reclamation opportunities.
Open Brand Mention Monitor →