SOLIDWEB / ARTICLE · 21.11.2026

Should businesses delete negative comments on social media?

Negative comments should generally be answered, not removed. Here is a practical model for choosing between responding, hiding, deleting and escalating.

Svar, skjul eller slett: Tiltaket må passe til kommentaren.

A negative comment can feel bigger than it is, especially when it appears publicly under the company’s name and someone has already had time to leave an enthusiastic question mark. The first impulse may be to delete it. That is often a mistake. Removing legitimate criticism is roughly like turning off the smoke alarm because the sound is unpleasant: The problem does not necessarily disappear, but you lose a useful signal.

The main rule is simple: Legitimate criticism should normally remain visible and receive a response. Comments containing spam, attempted fraud, personal data, threats or harassment can be hidden, deleted or escalated. Between these extremes lie the grey areas, and that is where a business needs a consistent assessment model.

The aim of moderation is not to make a comment section look like a brochure. The aim is to protect users, keep the conversation relevant and show that the business handles criticism in an orderly way.

Distinguish between criticism and rule-breaking

Start with the content of the comment, not with how uncomfortable it is to read. A dissatisfied customer can be sharp in tone without breaking any rules. Conversely, an apparently friendly comment may contain a scam link or collect contact details.

Use four categories when assessing a comment:

  • Legitimate criticism: Complaints about a product, delivery, customer service, price or the company’s actions. The comment should normally remain visible and receive a response.
  • False claims: Claims that can be corrected with verifiable facts. Respond briefly and precisely rather than deleting the comment, as long as it otherwise follows the rules.
  • Unwanted content: Spam, repeated advertising, fake competitions, suspicious links or comments with no relevant connection to the post. This can usually be removed.
  • Harmful or illegal content: Threats, hate speech, harassment, publication of personal data or other serious content. Document the comment, limit its visibility and follow the company’s escalation procedure.

Do not leave the entire assessment to a tool promising automatic sentiment analysis, crisis alerts and “reputation orchestration” in the same dashboard. Such systems can help with sorting, but irony, dialect and context remain challenging. A green smiley face next to an enraged comment is, in practice, just a digital mood ring with a subscription.

When should you respond, hide or delete?

Respond when the criticism is relevant

If the comment describes a genuine experience or asks a legitimate question, it should usually be answered publicly. The response is not just for the person who wrote the comment. It is also for everyone who reads the conversation later.

Kommentarer sorteres etter innhold og risiko, ikke ubehag.
Comments should be sorted by content and risk, not discomfort.

A good initial response should do three things: acknowledge the issue, explain what you can clarify publicly and show the next step. Avoid discussing personal details, order numbers or customer history in the comments. Instead, ask the person to continue the dialogue through a suitable private channel when the matter requires identification.

A practical response can be structured as follows:

  1. Thank the person for raising the issue.
  2. Confirm specifically what you have understood.
  3. Briefly explain what you can investigate or do.
  4. Explain how the person can contact you without publishing personal data.

Avoid standard responses such as “We’re sorry you feel that way.” The wording often sounds as though the problem is really the customer’s feelings. A response can be linguistically polished like an airport floor and about as warm.

Hide when visibility should be limited

Hiding may be appropriate for spam, repeated provocation, suspicious links or content that disrupts the conversation without requiring complete deletion. The function works differently across platforms. On some services, the comment may still be visible to the sender or parts of the sender’s network. Therefore, check what the function actually does before basing your moderation procedure on it.

Hiding should not be used as a discreet way to remove legitimate criticism. If the person or others discover this, the discussion may move to a channel you do not control. Screenshots also have a particular ability to outlive any cleanup.

Delete when the content clearly violates the rules

Deletion is appropriate when a comment contains fraud, spam, exposed personal data, serious harassment or other clear rule violations. Take a screenshot and note the time, account and reason before deleting it if the matter may require follow-up.

In the event of credible threats or serious security incidents, the comment should not be handled as ordinary customer dialogue. Preserve the necessary documentation, notify the responsible manager and follow the company’s emergency procedures. Any further reporting must be assessed based on the content and situation.

Create a moderation procedure that works during business hours

A good procedure must be short enough to be used in practice. If a comment has to go through the marketing manager, legal department, managing director and a status meeting next Thursday, even a simple response will be left to mature like yoghurt.

Assign responsibilities in advance:

En kort modereringsrutine gir raskere og tryggere oppfølging.
A short moderation procedure enables faster and safer follow-up.
  • First line: Reads comments, responds to common questions and categorises criticism.
  • Subject matter expert: Checks facts in matters concerning products, delivery, agreements or technical issues.
  • Communications manager: Handles matters with potential reputational risk or significant reach.
  • Management or emergency response lead: Make decisions in response to threats, security incidents or serious allegations.

Also define when frontline staff can respond without approval. Ready-made response templates can be useful for opening lines, contact information and privacy, but they should be adapted to the case. A template is a guardrail, not a remote control for human communication.

Set an internal response-time target based on staffing and channel. Do not promise round-the-clock follow-up if no one monitors the channel in the evening. Instead, clearly state when the channel is staffed, and set up alerts for comments containing certain risk-related terms. Automated alerts can support the work, but someone must still understand what they are alerting you to.

Write a short, visible moderation policy

A moderation policy explains what is allowed on the company’s own pages, and what may be removed. It should be written in plain language, not like a minor doctoral thesis on rules of conduct.

The policy should cover at least:

  • objective criticism and disagreement
  • personal attacks, harassment and hateful speech
  • threats and calls to cause harm
  • personal data and confidential information
  • spam, attempted fraud and unsolicited advertising
  • repeated posts that disrupt the conversation
  • possible measures, such as responding, hiding, deleting, blocking and reporting

The rules must be applied consistently, even when the criticism concerns an issue the company would rather not have to deal with. Moderation loses credibility if praise is left untouched while critical questions disappear conspicuously quickly.

Keep a simple moderation log for difficult cases. Note the date, channel, assessment, action taken and who made the decision. You do not need to build your own “social governance command center” with twelve statuses and purple labels. A well-organised, access-controlled form will often meet the need.

The practical conclusion is therefore: Do not delete a comment because it is negative. Delete or hide it because it breaches clear rules. Respond to legitimate criticism, move personal matters to a secure channel and document serious incidents before restricting content. That way, the comment section becomes neither an advertising billboard nor an unattended back room, but a channel the company can genuinely stand behind.