A name match is a lead, not proof of relevance. Check the product, domain, location and original context, and keep uncertain results visible.
A monitoring sheet contains dozens of matches for the same name. Some refer to another company; others use an ordinary word. A few cannot be identified from the headline. Counting them all distorts the report, while a broad exclusion may hide a real complaint.
Resolve which entity a mention concerns before assigning follow-up. This manual review method is designed for public material the team is entitled to access.
Build an alias list that can identify the business
Pair each name with identifying context. Record full names, abbreviations, former names and language variants alongside official domains, principal products, operating locations and confirmed official accounts. The context helps distinguish a business from a word match.
Keep known unrelated names in a separate list with an example and a reason. In a hypothetical case where a brand shares a city’s name, that location alone should not automatically exclude a post that also names the brand’s product or domain. Leave unconfirmed agents outside the official account list.
| Entry | Information to keep | Purpose |
|---|---|---|
| Brand names | Full, short and language variants | Find candidates |
| Identifiers | Domain, product, location, official account | Support attribution |
| Unrelated names | Other entity and original example | Recognise false matches |
| Unconfirmed relationships | Agents, old brands or similar names | Retain for review |
Read the original and use three outcomes
A headline match does not establish attribution. Open the original link and look for product names, location, legal entity and linked domains. Note the evidence supporting the decision. If the source has disappeared or requires access the team does not have, do not reconstruct missing facts.
Use relevant, unrelated and pending categories. An unrelated entry should identify the other entity where possible; a pending entry should state what is missing. A complaint that cannot yet be attributed can still carry a question for the business owner rather than disappearing from the sheet.
Test exclusions against known examples
Google Search documents quotation marks for exact phrases, site: for domains and the minus sign for exclusions. These are Google Search operators, not universal monitoring syntax. Check the capabilities of any other tool before applying them.
Compare the revised query against known relevant and irrelevant examples. If an exclusion hides a genuine complaint, remove or narrow it. Fewer results alone do not demonstrate better accuracy; the sample must show which useful material was retained.
Source: Official guidance
Hand over the reasons as well as the rule
Keep the original URL, decision reason, reviewer and rule version. A new reviewer can then reverse an incorrect exclusion and update the list after a rebrand, product launch or market change. Describe source visibility rather than presenting search results as a total count of online mentions.
Google Alerts offers language, region and source-type settings and sends notifications when it finds matching search results. That does not establish complete coverage or guaranteed detection. List inaccessible or unchecked channels explicitly.
| Original URL | Decision | Evidence or gap | Next action |
|---|---|---|---|
| Record each item | Relevant / unrelated / pending | Product, domain, context or missing detail | Keep, exclude or ask the business owner |
Source: Official guidance
To review persistent same-name noise, bring the alias list, positive and negative examples and current filters to PowerFront and agree a public-source review scope.