Table of contents
Two extremes appear in the same situation. One team treats every complaint as a crisis and stops all other work. Another keeps calling a spreading discussion "an individual user's gripe". Both misjudge for the same reason: no grading standard, so severity is decided by instinct.
1. Four Levels
| Level | Signals | Response window | Owner |
|---|---|---|---|
| Blue | Single complaint, no spread | Reply within 24 h | Support |
| Yellow | Same issue repeated, screenshots circulating | Statement drafted within 8 h | Marketing lead |
| Orange | Reaches industry groups, media enquiries | Unified line within 4 h | Brand lead |
| Red | Mainstream coverage or regulator attention | Immediate, executive level | Leadership |
Two variables drive the level: spread and whether safety or compliance is involved. Any question touching personal safety, falsified test data or qualification issues escalates one level regardless of volume.
2. Fields of a Response Record
A response record makes the process traceable. Fixed fields: time detected, platform, original link, spread estimate, core allegation, fact-check result, owning department, approved statement version, action taken, review date.
The field most often filled in badly is "core allegation". Teams record emotion ("users are angry") rather than a factual claim ("the customer says delivery was twenty days late with no notice"). Emotion cannot be checked; a factual claim can, and the fact-check result drives every subsequent action.
3. Five Steps
Monitor. Cover the brand name, product models, executive names, common misspellings, and industry terms combined with regions.
Assess. Separate factual error, experience problem and coordinated attack. Factual errors need evidence-based correction, experience problems need remedy and improvement, coordinated attacks go through platform rules.
Respond. Verify internally before speaking publicly. A fast reply on unverified facts usually creates a second wave.
Repair. Publishing the improvement works better than repeated explanation. Users care whether the problem is solved.
Review. Write this allegation into the next round of product documentation and FAQs so the same question does not recur.
4. What Requests Are Not Legitimate
Companies often define the goal as removing negative content. That request does not hold up: platform content is controlled by the platform, and any service promising removal carries compliance risk and should never appear in a contract.
Three workable directions instead. Where content is factually wrong, file a platform appeal with evidence. Where the experience problem is real, respond publicly with a remedy. Where search results look poor, build positive content over time — that is content work, not response work.
5. Three Numbers to Review
Repeat rate of the allegation, time to first unified statement, and secondary spread after the response. All three falling means the process is working.
Search boxes also return concatenated strings such as "online opinion opinion response", which shows the naming is unsettled. The name does not matter; the grading standard and the fact check are the reusable parts.
6. FAQ
Does a small company need this? Yes, in simplified form: one person monitoring, one record template, three levels.
Should the response name individuals? Name the role and the corrective action, never personal details.
How detailed should the process be? Detailed enough to say who does what within how many hours. Anything longer goes unread on the day.
Related
- See also: mapping a promotion plan to the decision chain
- See also: the website as an information base
GrowLab focuses on integrated content marketing, offering brand strategy, website development and search marketing (SEO/GEO) as a full-chain service. Contact us for an assessment of your own promotion plan.
Need a professional growth plan?
Add our WeChat for a tailored strategy and annual quote.
Contact Us