Which mentions are actually about your brand?
Brand names can be ordinary words, shared handles or abbreviations used by unrelated organizations. A mention count without disambiguation can be misleading before any analysis begins. Use a bounded public research workflow to establish which records refer to your brand, which need review and which are outside the selected population.
Review mentions of our brand in the selected accounts this month. Use our official domain to disambiguate namesakes, separate owned posts from independent mentions, and list ambiguous matches and source gaps.Write an identity rule
List the official domain, profile URLs, product names and spelling variants that identify your brand. Add explicit exclusions for namesakes. A match on a short word alone may be too weak; require contextual evidence such as the product category or a linked official account. Keep uncertain references separate instead of silently accepting them.
Choose the observation scope
State the platforms, communities or public accounts in the research brief, with the date interval and language where relevant. Start by checking actual source coverage. Search results, retained watchlist records and a full platform-wide population are different scopes. Do not present a selected-account review as exhaustive social monitoring.
Read mentions in context
Separate owned announcements, independent references and replies that mention another subject. Link the original record and distinguish a direct mention from an analyst's inferred association. Avoid automated emotional judgments where context is missing. Keep useful counterexamples so your team can see where the identity rule produces false matches.
Report evidence rather than a universal score
Deliver themes with the actual number of accepted records and ambiguous exclusions in the inspected sample. If you compare periods, hold the identity rule and account population constant. A source that was unavailable in one interval should not appear as a fall in brand interest. Any wider market conclusion needs evidence beyond the observed mentions.
Keep the boundaries clear.
- Selected records are not a complete count of every brand mention.
- Do not turn unavailable source coverage into a zero-mention period.
Start with a read-only check: ask your agent to report actual source coverage before collecting records or exporting an audience.
Primary references
Documentation reviewed 2026-11-13. Product capabilities, client availability and plans can change. Third-party marks identify their products, not an affiliation.