32 ChatGPT prompts for creators to turn hate comments into engaging content
Hate comments are content briefs that angry people hand you for free. Most creators treat them like spam and delete. The smart ones in 2026 treat them like research.
I wrote this guide because I got tired of watching brilliant creators leave the juiciest story of their week sitting in their notifications. The meanest comment in your inbox is usually the most honest one. It points at a misunderstanding, a missing story, or a nerve you accidentally hit. All three are content gold if you know what to do with them.
This article gives you 32 ready-to-paste ChatGPT prompts for turning hate comments into content, plus the framework I use to decide which comments are worth a post, which deserve a reply, and which deserve to die in a Notion graveyard.
You will also get a 14-day sprint, a People Also Ask section, a comparison table of prompt categories, and a list of common mistakes. Every stat is from 2026. Every prompt is multi-line, copy-pasteable, and built to ship.
Pull quote: “Posts that generate mid-length comments (50–99 characters) earn 151.6% higher engagement on average.” - Hootsuite Social Trends 2026 Report, published January 2026
Quick answer / TL;DR
- A hate comment is a free focus group. It tells you exactly what your audience is confused about, afraid of, or hungry to hear you say.
- The 4-part framework is Mine, Mirror, Make, Multiply. You mine for the underlying emotion, mirror it back without defensiveness, make a new piece of content, then multiply it across platforms.
- Replies that spark conversation lift engagement dramatically. According to Hootsuite’s 2026 data, proactive brand comments that get a reply earn 1.6x more engagement. Posts that get mid-length comments earn 151.6% more engagement.
- Comments are rising everywhere in 2026. LinkedIn comments are up 37% year over year and X replies jumped 107% year over year (Hootsuite, January 2026).
- Use ChatGPT to do the boring part (sort, summarize, paraphrase) so you can do the human part (decide, empathize, ship).
Why your haters are your next breakout post
A hate comment is a gift. It is the only message in your day that arrives pre-loaded with emotion, a specific objection, and a real human who cares enough to type. Most creators ignore that.
Here is the part that should change your mind: comment-heavy posts actually perform better. In the 2026 Hootsuite Social Trends report, posts that triggered mid-length comments (50–99 characters) earned 151.6% more engagement on average than posts with shorter or longer comments. And Hootsuite’s engagement benchmark report (May 2026) confirmed that good engagement rates now sit between 1.3% and 3.5% across the major platforms, with comment quality being one of the strongest signals of long-term reach.
Add to that the fact that LinkedIn comments are up 37% year over year, and the average number of replies per X post jumped 107% year over year (from 1.64 in 2023 to 3.4 in 2024, with continued growth into 2026). Comments are no longer a side metric. They are the metric.
So if your comments are getting meaner, congratulations. Your audience is getting bigger and more invested. The question is whether you treat that as a problem or as a content engine.
Creator reality check: the 2026 Pew Research survey of 5,022 U.S. adults found that 84% of U.S. adults use YouTube, 71% use Facebook, and 50% use Instagram. Half of all U.S. adults under 30 use Instagram (Pew Research, November 2025). The pool of people who can comment on your work is enormous. The slice who do is small. Listen to them.
The 4-part hate-to-content framework
Before you paste a single prompt into ChatGPT, lock in the framework. Every prompt below maps to one of these four moves. If you skip the framework, the prompts will produce generic AI mush. If you hold the framework, they will sound like you.
1. Mine
Mining is the act of pulling the actual concern out of the insult. Most hate comments are not about you. They are about a worry the commenter has. “This advice is mid” usually means “I tried this and it did not work for me.” “You sound like every other guru” usually means “I am tired of being sold to.”
The job in the Mine step is to surface that real concern so you can address it in public.
2. Mirror
Mirroring is acknowledging the comment without agreeing, deflecting, or scolding. You repeat back what you heard, in their words, in one sentence. This is straight out of Crucial Conversations by Patterson, Grenny, McMillan, and Switzler. The model works for creators the same way it works for hostage negotiators.
A good mirror takes 1–3 sentences. It should not be sarcastic. It should not be a dunk. It is a simple reflection: “Sounds like my last video left you feeling talked down to. That tracks with what I was trying to do.”
3. Make
Make is the content step. This is where you take the mined concern and the mirrored reply and turn them into a new post, video, newsletter, or carousel. The best Make outputs answer the mined concern better than the original comment did, and they do it for the silent majority who had the same thought but did not type it.
4. Multiply
Multiply is the distribution step. You take the new content and reshape it for every platform you touch. A YouTube reply becomes a TikTok, a LinkedIn post, a tweet thread, and a newsletter paragraph. Multiply is where ChatGPT earns its subscription, because reshaping is exactly the kind of work large language models are good at.
A note on how I built these prompts
A few ground rules I followed for every prompt below:
- I use multi-line prompts with explicit role, context, task, constraints, and output format. Single-line prompts are how you get vague answers.
- I lean on frameworks the model can copy. StoryBrand, JTBD (Jobs To Be Done), Nonviolent Communication by Marshall Rosenberg, the Hook-Story-Offer structure from the Pole Position playbook by Sam Parr, and the Rogan-style long-form debate style used by Lex Fridman.
- I include placeholders like
[PASTE_COMMENT]and[YOUR_NICHE]so you can swap in your own context without rewriting. - I prefer 2026 tool names (ChatGPT, Buffer, Hootsuite, Sprout Social, Agorapulse, Loom, Notion, Substack, Figma, Canva, Otter, Fathom) so the outputs feel current.
- I never use “delve,” “tapestry,” “in the ever-evolving landscape,” “moreover,” “furthermore,” or “it’s important to note.” You will not find them in this article either.
Comment-mining prompts (prompts 1-5)
These prompts help you turn a wall of mean DMs into a clean list of audience concerns you can address in public.
Prompt 1: Classify hate comments by underlying concern
Purpose: Sort 50+ comments into 5-7 underlying concerns in seconds, so you know which ones to address and which ones to ignore.
ROLE: You are an expert audience researcher trained in Jobs To Be Done (JTBD)
and Nonviolent Communication (Marshall Rosenberg).
CONTEXT: I am a creator in [YOUR_NICHE] with a community of
[AUDIENCE_SIZE] on [MAIN_PLATFORM]. Below is a dump of 50 recent
comments and DMs, including negative and hateful ones.
TASK:
1. Read every comment.
2. Cluster them into 5-7 underlying concerns. Examples of concerns
might be: "feels sold to," "disagrees with my politics," "thinks
I am out of touch," "wants more depth," "thinks I am too niche."
3. For each concern, list:
- A 1-3 word label
- The exact quotes that fit (3-5 examples)
- The percentage of comments that fit (rough estimate)
- A "silent majority" read: what non-commenters might be thinking
4. Flag the 3 concerns that would make the strongest content if I
addressed them publicly.
CONSTRAINTS:
- Do not soften the concerns. Real language only.
- Skip any comment that is pure slur or spam. Note "X% were pure
abuse with no usable signal" at the end.
INPUT: [PASTE_50_COMMENTS_HERE]
OUTPUT FORMAT:
Concern 1: [Label]
- Quotes: ...
- % of comments: ...
- Silent majority read: ...
- Worth addressing publicly? Yes / No
...
Summary: X% were pure abuse with no usable signal.Example output (abbreviated):
- Concern 1: “Feels sold to” (32% of comments). Quotes: “another plug,” “this is an ad.” Silent majority read: they enjoy the content but resent the pitch. Worth addressing? Yes.
- Concern 2: “Thinks I am out of touch” (18%). Quotes: “easy to say from your [lifestyle].” Silent majority: wondering if my advice applies to their reality.
Pro tip: Paste the same output into Notion as a single database row per concern. Tag each row with the platform. You now have a content calendar that you did not have to brainstorm.
Prompt 2: Translate a single hate comment into a usable insight
Purpose: Take one nasty comment and extract the story the commenter is actually telling.
ROLE: You are a JTBD interviewer who specializes in turning customer
objections into product insights.
CONTEXT: I received this comment on my [PLATFORM] post about
[POST_TOPIC]. It stung, but I want to use it.
COMMENT: "[PASTE_COMMENT]"
TASK:
1. Translate the comment from "what they said" to "what they mean."
2. List 3 possible jobs they are trying to do (Jobs To Be Done).
3. List 3 emotional drivers behind the comment.
4. Identify the one factual error or missing context in the comment,
if any. If there is none, say so.
5. Suggest 3 short content hooks (1 sentence each) that would
address the underlying concern without quoting the comment
back at them.
CONSTRAINTS:
- Be kind to the commenter. They are a real person.
- No sarcasm. No dunk.
- The 3 hooks must be platform-native to [PLATFORM] (e.g., a hook
on TikTok reads differently than a hook on LinkedIn).
OUTPUT FORMAT:
Translation: ...
Possible jobs: ...
Emotional drivers: ...
Factual check: ...
Hook 1 (for [PLATFORM]): ...
Hook 2: ...
Hook 3: ...Pro tip: Keep the original comment in a private Google Doc. Date it. Re-read it in 90 days. You will be shocked at how often the “hater” was right.
Prompt 3: Build a 30-comment “comment cloud” of recurring words
Purpose: Get a quick word-frequency read across a big comment dump to find themes you would have missed.
ROLE: You are a discourse analyst who uses simple text analysis.
CONTEXT: I have 100 recent comments on my [PLATFORM] post about
[TOPIC]. Many are negative or skeptical.
TASK:
1. Read all 100 comments.
2. Produce a "comment cloud" sorted by frequency. Group words into
4 buckets:
- Emotion words ("frustrated," "tired," "angry")
- Objection words ("doesn't work," "out of touch," "generic")
- Identity words ("as a mom," "as a dev," "in 2026")
- Question words ("how," "what about," "why")
3. Highlight the top 5 words in each bucket.
4. Pull 3 quotes that match the top emotion word.
5. Suggest one post angle that would address the most common
emotion and objection at the same time.
OUTPUT FORMAT:
Comment cloud (top 5 per bucket):
Emotion: ...
Objection: ...
Identity: ...
Question: ...
Quotes (top emotion): ...
Suggested post angle: ...Pro tip: Run this every Friday. You will start to see the same three objections show up across platforms. That is your pillar content.
Prompt 4: Map a comment thread into a StoryBrand problem-solution arc
Purpose: Turn a 30-reply comment thread into a one-paragraph StoryBrand-style narrative you can use as a video or post opening.
ROLE: You are a StoryBrand copywriter (Donald Miller framework) who
also understands creator content.
CONTEXT: I have a comment thread from a recent post. The thread is
attached below. It includes the original post caption, then 30
replies.
THREAD: [PASTE_THREAD]
TASK:
1. Identify the "Character" (the commenter or the silent viewer).
2. Identify the "Problem" they are voicing (external, internal,
philosophical).
3. Identify the "Guide" (me, the creator) and the empathy + authority
I need to show.
4. Identify the "Plan" (3 simple steps that would resolve the
problem).
5. Identify the "Call to Action" (one specific thing the viewer
should do).
6. Write a 1-paragraph StoryBrand version of this thread, around
80-120 words, that I can use as a video or post hook.
CONSTRAINTS:
- The paragraph must read like a human, not a brand deck.
- Use 2026 vocabulary (no "in today's fast-paced world").
- No hashtags in the paragraph.
OUTPUT FORMAT:
Character: ...
Problem: ...
Guide: ...
Plan: ...
CTA: ...
1-paragraph StoryBrand rewrite: ...Pro tip: StoryBrand reframes the comment from a complaint into a hero’s journey. That shift makes your reply land softer and your next post hit harder.
Prompt 5: Cluster “hate” comments into 4 audience segments
Purpose: Stop treating your audience as one block. Break the comment pool into 4 segments with different concerns.
ROLE: You are an audience segmentation strategist with a background
in community-led growth.
CONTEXT: Below are 40 negative or skeptical comments from my recent
[PLATFORM] post. My current audience size is [SIZE] in [NICHE].
COMMENTS: [PASTE_40_COMMENTS]
TASK:
1. Cluster the comments into exactly 4 segments. A segment should
represent a coherent worldview, not a demographic. Example
segment names: "The Burned Buyer," "The Free-Tier Skeptic,"
"The Platform Native," "The Old Guard."
2. For each segment, give:
- A 2-3 word name
- The 1-sentence worldview
- The 3 most representative quotes
- The 1 content idea that would resonate with them
- The 1 content idea that would alienate them
3. Identify which segment is loudest in the comments but smallest
in the broader audience (the "vocal minority").
4. Identify which segment is quietest in the comments but biggest
in the broader audience (the "silent majority").
5. Recommend whether to address the loudest or the quietest segment
in my next post, and why.
OUTPUT FORMAT:
Segment 1: ...
Quotes: ...
Content that resonates: ...
Content that alienates: ...
...
Strategic recommendation: ...Pro tip: This prompt reveals that the loudest haters are often a tiny slice of your real audience. Once you see that, you stop writing for them and start writing for the silent majority.
Reply and moderation prompts (prompts 6-11)
These prompts help you respond to comments faster, more kindly, and with less emotional bleed. Use them inside ChatGPT, then paste the result back into the platform.
Prompt 6: Generate a “mirror” reply using Crucial Conversations framing
Purpose: Write a 1-3 sentence reply that acknowledges the commenter’s concern without being defensive, using the mirror technique from Crucial Conversations by Patterson et al.
ROLE: You are a communication coach trained in Crucial Conversations
(by Patterson, Grenny, McMillan, and Switzler) and Nonviolent
Communication (Marshall Rosenberg).
CONTEXT: I am a [NICHE] creator. A follower left this comment on my
post about [POST_TOPIC]:
COMMENT: "[PASTE_COMMENT]"
TASK:
1. Write 3 reply options. Each reply must:
- Mirror back the concern in their words (not mine)
- Contain zero defensiveness
- Avoid sarcasm, dunking, or guilt
- Open a door to a follow-up reply
2. Keep each reply under 280 characters.
3. After the replies, list 3 follow-up replies I could write if they
respond.
4. Flag the reply that is most likely to spark a thoughtful
conversation (not just a like).
CONSTRAINTS:
- No emoji in the reply itself.
- Do not mention the comment is "hateful" or "negative." That
escalates.
OUTPUT FORMAT:
Reply A (mirror): ...
Reply B (mirror + question): ...
Reply C (mirror + tiny concession): ...
Most conversation-sparking: Reply [X]
Follow-up if they engage: ...Pro tip: Save your three favorite replies as a “Mirror Bank” in a Notion page. Reuse the structure every week. The more consistent your mirrors, the more your audience learns how to engage with you.
Prompt 7: Convert a hot reply into a calm public answer
Purpose: Take an emotionally charged comment and turn it into a calm, public, teachable reply.
ROLE: You are a customer-support lead who specializes in de-escalation.
CONTEXT: A comment on my [PLATFORM] post has gone sideways. It
started as a disagreement and is now personal.
COMMENT THREAD: [PASTE_THREAD]
TASK:
1. Identify the emotional escalation point.
2. Write a public reply that:
- Acknowledges the emotion ("This clearly matters to you")
- States a fact that cannot be argued with
- Reframes the disagreement as a shared value
- Offers a path forward (a question, a link, a DM)
3. Keep it under 200 words.
4. Add a private DM template I can send if the thread goes toxic.
OUTPUT FORMAT:
Public reply: ...
Private DM template: ...
Why this works: ...Pro tip: Post the public reply and then wait. The community will often pile on with support, which does more for your brand than any post you could write.
Prompt 8: Auto-tag a comment as “engage / archive / block”
Purpose: Triage 100 comments in minutes so you spend your energy on the ones that matter.
ROLE: You are a community manager with 10 years of moderation
experience across Instagram, TikTok, YouTube, and LinkedIn.
CONTEXT: Below are 100 recent comments on my posts. Sort each into
one of three buckets:
- ENGAGE: deserves a thoughtful reply (has a real question, a real
objection, or a thoughtful disagreement)
- ARCHIVE: do not reply, but keep in the swipe file (mild snark,
off-topic, generic "great post")
- BLOCK: delete or hide (slur, spam, doxx, threat, harassment)
TASK:
1. Tag every comment with one of the three labels.
2. For ENGAGE comments, write a 1-sentence reply suggestion.
3. For ARCHIVE comments, note 1 word that describes the vibe
(e.g., "snark," "shill," "lazy").
4. For BLOCK comments, list the trigger word or pattern so I can
build a moderation filter.
5. End with a one-line summary: "X% ENGAGE, Y% ARCHIVE, Z% BLOCK."
COMMENTS: [PASTE_100_COMMENTS]
OUTPUT FORMAT:
1. [ENGAGE] "Comment text..." -> Suggested reply: ...
2. [ARCHIVE] "Comment text..." -> Vibe: ...
3. [BLOCK] "Comment text..." -> Trigger: ...
...
Summary: X% / Y% / Z%Pro tip: Run this once a week. The percentages will start to reveal which topics pull ENGAGE and which pull BLOCK. Lean into the ENGAGE topics.
Prompt 9: Generate a “kindly firm” reply for a personal attack
Purpose: Respond to a personal attack without apologizing for things you did not do, and without escalating.
ROLE: You are a communication coach and a former journalist trained
in de-escalation.
CONTEXT: I received this comment that crosses the line from
criticism to personal attack:
COMMENT: "[PASTE_COMMENT]"
TASK:
1. Write 2 reply options:
- Option A: a public reply that sets a boundary without naming
the attack. Use a calm, "I" voice.
- Option B: a private DM that addresses the comment directly and
invites a reset.
2. For each option, explain in 1 sentence why it works.
3. Recommend a moderation action (delete, hide, block, ignore).
4. Suggest 1 sentence I can use in a future post about handling
online negativity, attributed to the situation, not the user.
CONSTRAINTS:
- Never insult back.
- Never reveal personal details.
- Never use "we" if I am a solo creator. Use "I."
OUTPUT FORMAT:
Option A (public): ...
Why it works: ...
Option B (DM): ...
Why it works: ...
Recommended moderation action: ...
Reusable line for future content: ...Pro tip: Most personal attacks land because the commenter’s expectation is silence or a fight. A calm “I” reply throws off the script. It also reads well to everyone else watching.
Prompt 10: Generate a “thanks, I missed that” reply for a fair critique
Purpose: Turn a fair critique into a public demonstration of humility, which boosts trust.
ROLE: You are a brand strategist trained in story-based marketing
and the StoryBrand framework.
CONTEXT: A long-time follower left a thoughtful critique on my post
about [TOPIC]. They are right about part of it.
COMMENT: "[PASTE_COMMENT]"
TASK:
1. Write a public reply that:
- Thanks them by first name or @ handle
- Names the specific thing they got right
- Acknowledges what I missed or under-explained
- Promises one small, specific change in the next post
2. Keep it under 150 words.
3. Add a one-line caption I could put on a follow-up Story or
short-form video: "When the audience is right, the move is
obvious."
CONSTRAINTS:
- No self-deprecation. No "I'm such a fraud."
- No "you're so right, I should learn from you" cringe.
OUTPUT FORMAT:
Public reply: ...
Follow-up Story caption: ...Pro tip: Audiences reward creators who update in public. A “thanks, I missed that” reply done right is more viral than the original post.
Prompt 11: Build a personal ban-list and reply-list
Purpose: Codify who is welcome in your comments and how you reply to each pattern.
ROLE: You are a community policy writer.
CONTEXT: I want to make my comment policy explicit so my future
replies stay consistent and my moderators know the rules.
TASK:
1. Write a 6-bucket comment policy for my [PLATFORM] community.
Buckets:
- Welcome (who thrives here)
- Welcome with friction (what I will reply to)
- Archive only (what I will not reply to)
- Hide (what I will hide from public view)
- Block (what gets a permanent block)
- Report (what I escalate to the platform)
2. For each bucket, give:
- A 1-sentence rule
- 2 example comments
- The moderation action
- A 1-line template reply, if any
3. Add a short "Why I moderate this way" note I can pin in my
bio or About page.
OUTPUT FORMAT:
Bucket 1: Welcome
Rule: ...
Examples: ...
Action: ...
Template reply (if any): ...
...
"Why I moderate this way" note: ...Pro tip: Publish this policy as a pinned post. Half your moderation load disappears when the rules are visible.
Reframe and rebuild prompts (prompts 12-17)
These prompts help you flip a hate comment into a new angle, a better frame, and a piece of content worth publishing.
Prompt 12: Reframe a hate comment as a “yes, and” creative brief
Purpose: Use improv’s “yes, and” rule to turn a cruel comment into a creative constraint.
ROLE: You are a creative director trained in improv comedy's "yes,
and" principle and the JTBD framework.
CONTEXT: A hater left this comment on my recent post:
COMMENT: "[PASTE_COMMENT]"
TASK:
1. Start with a literal "Yes, and..." reply that accepts the
comment's premise (even the unfair parts) and adds a constructive
layer.
2. Translate the comment into a creative brief with 3 constraints.
Example: "Make a 60-second TikTok that proves the opposite of
this comment, in my own style, without quoting them."
3. Brainstorm 3 video or post ideas that satisfy the brief.
4. For each idea, list:
- The hook (1 sentence)
- The body (3 bullet points)
- The CTA (1 sentence)
5. Pick the strongest idea and write a 100-word script.
CONSTRAINTS:
- Keep the commenter's identity private. No public callouts.
- The 3 ideas must be producible in under 90 minutes.
OUTPUT FORMAT:
Yes, and... reply: ...
Creative brief constraints: ...
Idea 1: ...
Idea 2: ...
Idea 3: ...
Chosen idea (full script): ...Pro tip: “Yes, and” works because it is the only move that does not waste your energy. You accept the comment, then redirect toward something useful.
Prompt 13: Build a “hate to hook” library for future content
Purpose: Convert a month of hate comments into a swipe file of 30 future hooks.
ROLE: You are a viral copywriter trained on creators like Lex
Fridman (long-form debate), Sam Parr (Pole Position hooks), and
modern LinkedIn creators.
CONTEXT: Below are 30 negative or skeptical comments I collected
this month. I want to convert them into future content, not feed
them.
COMMENTS: [PASTE_30_COMMENTS]
TASK:
1. For each comment, generate 1 hook (under 20 words) that:
- Could open a TikTok, Reel, Short, or LinkedIn post
- Does not name the commenter
- Captures the underlying tension in a single line
2. Group the 30 hooks into 5 themes. Give each theme a name.
3. For each theme, suggest a 3-post mini-series that would close
the loop on the underlying concern.
4. End with a "hook bank" formatted for paste-and-post use.
OUTPUT FORMAT:
Theme 1: [Name]
Hooks: 1. ... 2. ... 3. ...
Mini-series: Post 1 / Post 2 / Post 3
...
Hook bank:
- "Hook text 1"
- "Hook text 2"
- ...Pro tip: Every Friday, copy the “hook bank” into a Notion database. The next time you sit down to write, you will not start with a blank page.
Prompt 14: Find the question hiding inside the insult
Purpose: Extract the real question from a comment that reads like a personal attack.
ROLE: You are a Socratic teacher who can find the real question
behind any heated statement.
CONTEXT: A hater said this on my post:
COMMENT: "[PASTE_COMMENT]"
TASK:
1. Identify the explicit surface meaning.
2. Identify the implicit question hiding inside.
3. Rewrite the implicit question as a clear, neutral sentence.
4. Answer that neutral question in 3-5 sentences, in my creator
voice. Use a [NICHE] tone of voice.
5. Suggest a 30-second video or 1-tweet thread that delivers this
answer.
CONSTRAINTS:
- The answer must stand on its own without referencing the
comment.
- Avoid phrases like "as I mentioned" or "to address your concern."
OUTPUT FORMAT:
Surface meaning: ...
Hidden question: ...
Neutralized question: ...
Answer (3-5 sentences): ...
Short-form version: ...Pro tip: This is the prompt I use most often. It turns 80% of hate comments into teachable moments that build authority.
Prompt 15: Reverse-engineer a hater’s identity and intent
Purpose: Quickly figure out whether a hater is a lost fan, a competitor, a bot, or a real person with a real concern.
ROLE: You are an online community analyst trained in pattern
recognition and basic OSINT.
CONTEXT: A user I have never interacted with left this comment:
COMMENT: "[PASTE_COMMENT]"
USER HANDLE: [PASTE_HANDLE]
USER BIO (if available): [PASTE_BIO]
TASK:
1. Estimate, in a single sentence, the most likely identity:
- Lost fan (liked previous work, disagrees here)
- Curious skeptic (open-minded, testing the water)
- Trolling bad-faith actor
- Bot or spam
- Industry peer or competitor
- Real customer with a real problem
2. Estimate their likely intent: attention, validation, change,
status, sabotage.
3. Recommend a single best response action (reply publicly, DM
privately, ignore, hide, block).
4. Suggest one question I can ask in my reply that would confirm
the identity.
CONSTRAINTS:
- Do not assume malice. Default to "real person with a real
concern" unless the language clearly signals otherwise.
- Do not encourage doxxing or stalking.
OUTPUT FORMAT:
Most likely identity: ...
Likely intent: ...
Best response action: ...
Confirmation question: ...Pro tip: Most creators guess wrong here. They assume the loudest hater is a troll. It is usually a past customer who felt ignored. This prompt recalibrates your gut.
Prompt 16: Use Marshall Rosenberg’s NVC to write a non-defensive reply
Purpose: Apply Nonviolent Communication (observations, feelings, needs, requests) to write a reply that disarms.
ROLE: You are a certified Nonviolent Communication (Marshall
Rosenberg) practitioner.
CONTEXT: I received this comment on my post about [TOPIC]:
COMMENT: "[PASTE_COMMENT]"
TASK:
Write a 4-part reply using the NVC structure:
1. Observation: state the fact, no judgment.
2. Feeling: name the feeling you (or the commenter) might be
experiencing.
3. Need: identify the underlying need.
4. Request: make a specific, positive, doable request.
CONSTRAINTS:
- The reply must be under 100 words.
- Do not use "I feel that you" (a hidden judgment).
- Do not use "but."
- Do not use the word "should."
OUTPUT FORMAT:
Observation: ...
Feeling: ...
Need: ...
Request: ...
Combined reply: ...Pro tip: NVC is the only framework I have seen that consistently produces a reply everyone in the thread respects, including the silent majority.
Prompt 17: Convert a hater’s rant into a “what they got right” summary
Purpose: Pull the kernel of truth from a long rant, even if the rest is hostile.
ROLE: You are a fair-minded editor who can separate signal from
noise in any customer email.
CONTEXT: A long-time follower left this rant under my post:
COMMENT: "[PASTE_COMMENT]"
TASK:
1. Quote the 1-2 sentences that contain real signal (if any).
2. Quote the 1-2 sentences that are pure venting (so I know to
ignore them).
3. Summarize the real signal in 1 neutral sentence.
4. List 3 changes I could make to my content, product, or
community that would address the real signal.
5. Write a public reply that:
- Thanks them for the signal
- Names one change I will make
- Does not engage with the venting
6. Keep the reply under 120 words.
OUTPUT FORMAT:
Real signal quotes: ...
Venting quotes: ...
Neutral summary: ...
Actionable changes: ...
Public reply: ...Pro tip: Save the “actionable changes” list in a Notion database titled “Things I should have done sooner.” That database will quietly become the most valuable doc in your business.
New-content prompts from comments (prompts 18-23)
These prompts turn the framework outputs into ready-to-publish content. Each one targets a specific format.
Prompt 18: Write a 6-slide Instagram carousel from a hate comment
Purpose: Convert a single hate comment into a swipeable carousel.
ROLE: You are a designer-marketer hybrid who has shipped 100+
high-performing Instagram carousels. You use Canva and Figma for
mockups and ChatGPT for copy.
CONTEXT: This comment came in this morning:
COMMENT: "[PASTE_COMMENT]"
TASK:
1. Write copy for a 6-slide Instagram carousel that addresses the
underlying concern.
2. Slide 1 must be a hook (under 10 words, scroll-stopping).
3. Slides 2-5 must each deliver one distinct idea, with a bold
title and 2-3 lines of body copy.
4. Slide 6 must include a CTA: save, share, or comment.
5. For each slide, suggest:
- Background color (from a calm, modern 2026 palette)
- A simple Canva-ready visual concept (no stock photo needed)
- One bold title (under 8 words)
6. Add a 2-line caption for the post and 10 hashtag suggestions.
CONSTRAINTS:
- No "in today's world" cliches.
- No emoji in slide body copy. One emoji in the caption is fine.
OUTPUT FORMAT:
Slide 1
Title: ...
Body: ...
Visual: ...
Slide 2
...
Caption: ...
Hashtags: ...Pro tip: Carousels with a strong “save” CTA outperform “like” CTAs. The above template defaults to “save,” which is the highest-signal engagement metric in 2026.
Prompt 19: Write a 60-second TikTok script that answers a hater
Purpose: Convert a hater’s question into a fast, hooky TikTok.
ROLE: You are a short-form video scriptwriter in the style of
modern TikTok creators who use the Hook-Story-Offer structure
and the "Pole Position" hook style popularized by Sam Parr.
CONTEXT: A hater left this comment:
COMMENT: "[PASTE_COMMENT]"
TASK:
Write a 60-second TikTok script with the following structure:
- Hook (0-3 seconds): one sentence that names the objection.
- Beat 1 (3-15 seconds): the one piece of context most people
miss.
- Beat 2 (15-35 seconds): the practical reframe.
- Beat 3 (35-50 seconds): the specific example or story.
- CTA (50-60 seconds): a question to the audience that invites
replies.
ALSO INCLUDE:
- 3 on-screen text overlays (each under 8 words)
- 2 B-roll suggestions
- 1 caption for the post
- 3 alternative hooks to A/B test
CONSTRAINTS:
- Under 100 spoken words total.
- No hashtags in the script.
- No "let me explain" openers.
OUTPUT FORMAT:
Hook: ...
Beat 1: ...
Beat 2: ...
Beat 3: ...
CTA: ...
On-screen text: ...
B-roll: ...
Caption: ...
Alt hooks: ...Pro tip: Use the “alt hooks” in your next 3 TikToks. The fastest way to find a winning hook is to test variations, not to write a new post.
Prompt 20: Write a long-form Substack essay from a week of hate comments
Purpose: Convert a week’s worth of hate into a thoughtful 1,200-word essay.
ROLE: You are a long-form essayist in the style of Substack writers
who blend personal story with cultural critique.
CONTEXT: This week, my audience left these 8 comments and DMs that
represent the dominant pushback I received on [TOPIC]:
COMMENTS: [PASTE_8_COMMENTS]
TASK:
Write a 1,200-word Substack essay with this structure:
1. Cold open (1 paragraph): a specific scene that captures the
pushback.
2. The surface argument (1 paragraph): what the comments literally
say.
3. The deeper fear (1 paragraph): what the comments are really
worried about.
4. My former self (1 paragraph): the version of me that would have
agreed with the comments two years ago.
5. What changed (2-3 paragraphs): the data, the experience, the
conversation that shifted my view.
6. Where I am still wrong (1 paragraph): the part of the
criticism I have not yet figured out.
7. An invitation (1 paragraph): one specific thing I want the
reader to do or try.
CONSTRAINTS:
- No AI-tell words ("delve," "tapestry," "moreover," "furthermore,"
"in the ever-evolving landscape," "it's important to note").
- Use first-person ("I," "we," "you") throughout.
- Keep paragraphs under 4 sentences.
- End with a one-line CTA, not a "Like and subscribe" line.
OUTPUT FORMAT:
[Full essay, ready to paste into Substack]Pro tip: Substack readers reward nuance. The “Where I am still wrong” section is the part your readers will quote.
Prompt 21: Write a LinkedIn post that turns a workplace hot take into a lesson
Purpose: Convert a snarky LinkedIn comment into a high-engagement post.
ROLE: You are a B2B ghostwriter who has written 500+ high-engagement
LinkedIn posts. You understand that 2026 LinkedIn rewards
story-driven, contrarian-but-credible takes.
CONTEXT: A connection left this comment on a recent post of mine:
COMMENT: "[PASTE_COMMENT]"
TASK:
1. Write a 900-1,200 character LinkedIn post that:
- Opens with a one-line "I was wrong" or "I learned this the
hard way" hook
- Tells a 3-sentence story that includes the criticism
- Names the lesson
- Ends with a single, specific question
2. Add 3 alternative opening lines.
3. Add a DM template I can send to anyone who replies with
their own version of the story.
4. Recommend the best day and time to post, based on Hootsuite
2026 data, for a B2B [NICHE] audience.
CONSTRAINTS:
- 1-2 emoji max.
- No hashtag walls. 3 hashtags max, at the bottom.
- No "Agree?" as the closing question. Ask something specific.
OUTPUT FORMAT:
Post: ...
Alt opening 1: ...
Alt opening 2: ...
Alt opening 3: ...
DM template: ...
Best day/time: ...Pro tip: LinkedIn comments are up 37% year over year according to Hootsuite’s 2026 stats. Posts that invite a specific question get 2-3x more replies than posts that close with a generic “Thoughts?”
Prompt 22: Convert a hateful DM into a YouTube video title and thumbnail concept
Purpose: Take a single hateful DM and turn it into a click-worthy video that does not feel like a “response video.”
ROLE: You are a YouTube strategist who has shipped videos with
millions of views. You use Figma for thumbnail mockups and
understand 2026 YouTube retention curves.
CONTEXT: A subscriber DMed me this:
DM: "[PASTE_DM]"
TASK:
1. Write 5 possible YouTube video titles (each under 60 characters)
that would address the underlying concern without quoting the
DM.
2. Pick the best title. Explain why in 1 sentence.
3. Write a 30-second intro script for the video that:
- Acknowledges the DM indirectly
- Promises a specific takeaway
- Sets a curiosity loop
4. Describe 2 thumbnail concepts in detail:
- Layout (left vs. right)
- Text (under 4 words)
- Facial expression and pose
- Background color
- One prop or visual hook
5. List 3 YouTube chapters with timestamps.
6. Suggest a pinned comment that invites replies.
CONSTRAINTS:
- No "responding to hate" framing. The video should be a standalone
idea.
- No "cancel me" bait.
- Thumbnail text must be readable at 1x mobile size.
OUTPUT FORMAT:
Titles: ...
Best title: ...
Why: ...
Intro script: ...
Thumbnail 1: ...
Thumbnail 2: ...
Chapters: ...
Pinned comment: ...Pro tip: The best YouTube reply videos do not look like reply videos. They look like fresh ideas the audience did not know they needed.
Prompt 23: Write a podcast episode outline that addresses a controversial comment
Purpose: Use a controversial comment as the seed of a 30-minute podcast episode.
ROLE: You are a podcast producer trained in the Rogan-style
long-form debate style (Lex Fridman) and the StoryBrand framework.
CONTEXT: A listener left this comment:
COMMENT: "[PASTE_COMMENT]"
TASK:
Create a 30-minute episode outline with:
1. A working title (under 8 words).
2. A 1-sentence episode logline (the "what is this about").
3. A 3-act structure:
- Act 1 (0-5 min): the comment that started it
- Act 2 (5-20 min): 3 segments that explore the issue
- Act 3 (20-30 min): the synthesis and the open question
4. For each segment, list:
- A working title
- 3 talking points
- 1 quote or story to anchor it
5. List 3 guest archetypes who would be perfect for the episode
(do not name real people without permission).
6. Suggest 1 listener-submission question to feature in Act 3.
CONSTRAINTS:
- No "welcome to the show" intro plan. Keep it tight.
- 30 minutes total, with 2 mid-roll ad slots.
OUTPUT FORMAT:
Working title: ...
Logline: ...
Act 1: ...
Act 2:
Segment 1: ...
Segment 2: ...
Segment 3: ...
Act 3: ...
Guest archetypes: ...
Listener question: ...Pro tip: If the comment is from a real listener, DM them and ask if they want to come on the show. “Yes, we want to hear it in full” is a line that converts critics into evangelists.
Multi-platform post prompts (prompts 24-28)
These prompts help you take one piece of content and reshape it for every platform you touch, while keeping each post native.
Prompt 24: Repurpose a long post into a 7-platform content plan
Purpose: Turn one idea into 7 platform-native posts, scheduled across one week.
ROLE: You are a content repurposing strategist who has managed
multi-platform calendars for creators with 1M+ followers.
CONTEXT: I just published this long-form piece (Substack / YouTube
/ podcast / blog). It is about [TOPIC].
SOURCE PIECE: [PASTE_SUMMARY_OR_LINK]
TASK:
Create a 7-day, 7-platform content plan. For each day, output:
- Platform
- Post format (carousel, single image, text, video, thread, etc.)
- A complete draft of the post (under the platform's character limit)
- The recommended time to post, based on Hootsuite's 2026
engagement data
- A specific CTA tailored to that platform
PLATFORMS (in order):
1. LinkedIn (text post)
2. Instagram (carousel, 6 slides)
3. TikTok (60-second script)
4. X / Twitter (thread, 6-8 tweets)
5. YouTube (community post + Short script)
6. Substack Notes (short note + link)
7. Facebook (text post with link)
CONSTRAINTS:
- Every post must be native to the platform. No copy-paste.
- Every post must stand alone without the source piece.
- Use ChatGPT, Buffer, Hootsuite, Sprout Social, Agorapulse, or
Notion to manage the schedule.
OUTPUT FORMAT:
Day 1 - LinkedIn
Format: ...
Post: ...
Time: ...
CTA: ...
Day 2 - Instagram
...Pro tip: Repurposing is where the multiplication step pays off. One well-built long-form piece should easily produce 7 platform-native posts.
Prompt 25: Write a thread that uses a hater’s tweet as the opening hook
Purpose: Convert a single mean tweet into a thread that drives followers, not flame wars.
ROLE: You are a thread writer trained on the best of X / Twitter
in 2026. You know the difference between a viral thread and a
dunk thread.
CONTEXT: This tweet was quote-tweeted at me:
TWEET: "[PASTE_TWEET]"
TASK:
Write a 7-tweet thread that:
- Tweet 1: hooks with the criticism (paraphrased, not quoted)
- Tweet 2: acknowledges the surface argument
- Tweet 3: reframes with one new piece of data
- Tweet 4: tells a 1-sentence story
- Tweet 5: offers a specific, practical tip
- Tweet 6: invites a reply with a single question
- Tweet 7: links to one piece of long-form content (Substack, blog,
YouTube) that goes deeper
CONSTRAINTS:
- No quote-tweets of the original. Paraphrase only.
- No "well, actually" tone.
- No hashtag spam. 1-2 hashtags max, on the final tweet.
- Under 280 characters per tweet.
OUTPUT FORMAT:
1. ...
2. ...
...
7. ...Pro tip: Paraphrasing instead of quote-tweeting is the difference between a thread that grows your account and a thread that gets you suspended.
Prompt 26: Build a 5-Instagram-Story sequence from a single hot take
Purpose: Convert a single hot take into a 5-frame Story sequence.
ROLE: You are a designer-strategist hybrid who specializes in
Instagram Stories in 2026.
CONTEXT: I want to respond to this comment on my latest post:
COMMENT: "[PASTE_COMMENT]"
TASK:
Design a 5-frame Instagram Story sequence:
- Frame 1: a poll or question sticker that hooks attention
- Frame 2: a bold statement that names the criticism (paraphrased)
- Frame 3: a 2-sentence reframe
- Frame 4: a 1-sentence story or example
- Frame 5: a CTA sticker (question, quiz, or DM prompt)
FOR EACH FRAME, INCLUDE:
- Visual concept (1-2 sentences)
- On-screen text (under 12 words)
- Sticker type (poll, question, quiz, slider, link, etc.)
- Sticker copy
- Recommended background color from a 2026 palette
- A 1-line alt version for creators with smaller followings
(under 10K)
CONSTRAINTS:
- Each frame must feel like a native Story, not a slide.
- The sequence must work even if viewers only see frame 1 and
frame 5.
OUTPUT FORMAT:
Frame 1
Visual: ...
Text: ...
Sticker: ...
Sticker copy: ...
Color: ...
Alt version: ...
...Pro tip: Stories with question stickers consistently outperform static text-only Stories. Use a question on at least one frame, even if you also use a poll.
Prompt 27: Build a Notion content calendar from a week of hate comments
Purpose: Convert 7 days of hate comments into a 30-day Notion calendar.
ROLE: You are a Notion power user and a content strategist.
CONTEXT: Below are 30 hate or skeptical comments I collected this
week. I want to turn them into a 30-day content calendar in Notion.
COMMENTS: [PASTE_30_COMMENTS]
TASK:
1. Group the 30 comments into 5 content pillars.
2. For each pillar, write:
- A pillar name
- The underlying audience concern
- 6 post ideas (1 per day, Mon-Sat)
3. For each post idea, output a Notion database row with these
fields:
- Title (under 10 words)
- Platform (LinkedIn, Instagram, TikTok, X, YouTube, Substack)
- Format (text, carousel, video, thread, etc.)
- Hook (1 sentence)
- CTA (1 sentence)
- Source comment (paraphrased)
- Status (Default: Draft)
- Owner (Default: me)
4. End with the Notion CSV or Markdown table I can paste directly
into a database.
OUTPUT FORMAT:
Pillar 1: ...
Post 1: ...
Post 2: ...
...
Pillar 2: ...
...
Notion table:
| Title | Platform | Format | Hook | CTA | Source | Status | Owner |
| ... | ... | ... | ... | ... | ... | Draft | me |Pro tip: Use Notion’s database view to sort by status. Once you ship one, mark it “Posted” and add the URL. You will start to see which pillars earn the most engagement over time.
Prompt 28: Convert a hate comment into a community spotlight post
Purpose: Highlight a thoughtful hater, with their permission, to model great engagement.
ROLE: You are a community manager trained in the StoryBrand
framework and in creator-led community growth.
CONTASK:
1. Identify the comment in the list that is critical but
constructive.
2. Draft a public post that:
- Quotes the comment (with permission, paraphrased if needed)
- Names the specific insight they brought
- Credits them as a community member (with their consent)
- Invites other readers to add their version
3. Write a DM template to ask for the commenter's permission to
quote them.
4. Suggest 2 ways to follow up with the commenter to deepen the
relationship.
OUTPUT FORMAT:
Featured comment (paraphrased): ...
Public post draft: ...
Permission DM: ...
Follow-up plan: ...Pro tip: When you spotlight a thoughtful critic, the entire community recalibrates. It is the cheapest culture shift you can run.
Crisis and brand-safety prompts (prompts 29-32)
These prompts help you handle the worst comments: doxxing, threats, harassment, coordinated pile-ons, and full-blown crises. Use them carefully and consult legal counsel when needed.
Prompt 29: Decide whether a comment is a crisis, a complaint, or a vibe
Purpose: Classify a comment or thread in 30 seconds and pick the right response.
ROLE: You are a crisis communications expert trained on the
2025 Edelman Trust Barometer and modern creator-led crisis playbooks.
CONTEXT: Below is a comment or thread on my [PLATFORM] account.
Classify it into one of three categories:
- VIBE: standard negativity, ignore or light reply
- COMPLAINT: a real grievance, reply with empathy and a fix
- CRISIS: a coordinated attack, a doxx, a threat, or a viral
pile-on. Requires a formal response and a holding statement.
COMMENT OR THREAD: [PASTE_COMMENT_OR_THREAD]
TASK:
1. Classify the comment as VIBE, COMPLAINT, or CRISIS. Justify
the choice in 1 sentence.
2. For VIBE: suggest a 1-line public reply or "ignore."
3. For COMPLAINT: write a public reply, a private DM, and a fix
in 3 steps.
4. For CRISIS: produce:
- A 30-minute action plan (5 steps)
- A holding statement (under 100 words, public)
- A private statement for staff and moderators
- A moderation action list (which comments to hide, block,
report)
- A 24-hour follow-up plan
CONSTRAINTS:
- For CRISIS, do not delete comments blindly. Hiding with a
reason is safer than deleting.
- Never recommend doxxing, brigading, or paying off attackers.
- Recommend legal counsel for any threat or doxx.
OUTPUT FORMAT:
Classification: ...
Justification: ...
Public reply / plan: ...
...Pro tip: Most creators escalate comments from VIBE to CRISIS in their head. This prompt forces you to slow down and pick the right level of response.
Prompt 30: Build a one-page crisis comms doc for creators
Purpose: A reusable crisis plan you fill in once and update quarterly.
ROLE: You are a crisis communications strategist for solo creators
and small teams.
TASK:
Produce a one-page creator crisis comms template with these
sections:
1. Pre-crisis setup
- Pre-drafted holding statement
- List of moderators with phone numbers
- List of legal contacts
- Backup channel (Substack, personal email)
2. First 30 minutes
- What to do, in order
- What NOT to do
3. First 24 hours
- Public cadence (when to post)
- Internal cadence (when to update the team)
4. Aftermath
- How to debrief
- How to update the doc for next time
CONSTRAINTS:
- Designed for a solo creator or a team of 3.
- Plain language, no PR jargon.
- Use Hootsuite or Sprout Social for moderation queues.
OUTPUT FORMAT:
[Filled-in one-page template, ready to paste into Notion or Google Docs]Pro tip: Fill this in once. Update it every quarter. The day you need it, you will be too stressed to write it from scratch.
Prompt 31: De-escalate a coordinated pile-on without losing your audience
Purpose: Survive a coordinated attack with your reputation and your community intact.
ROLE: You are a crisis comms lead with 15 years of experience
handling pile-ons for public figures.
CONTEXT: My [PLATFORM] account is being targeted by a coordinated
pile-on. The trigger was [TRIGGER_EVENT_OR_COMMENT].
TASK:
Write:
1. A short public holding statement (under 100 words). Tone: calm,
factual, accountable, no defensiveness.
2. A pinned comment under the offending post, if applicable.
3. A private Slack message to my team, under 200 words.
4. A 24-hour follow-up message that I can post if the pile-on
continues.
5. A list of 5 moderation actions to take in the first hour.
6. A 3-sentence line I can use in a future post, six months from
now, that closes the loop on this incident.
CONSTRAINTS:
- Do not name the attackers.
- Do not escalate the original trigger event.
- Do not promise more than I can deliver.
- Acknowledge the 2025 Edelman Trust Barometer insight that
audiences trust individuals over brands in a crisis.
OUTPUT FORMAT:
Holding statement: ...
Pinned comment: ...
Team Slack message: ...
24-hour follow-up: ...
Moderation actions: ...
Future loop-closer: ...Pro tip: The “future loop-closer” line is the most important deliverable here. Six months from now, the audience will remember how you closed the chapter, not how you opened it.
Prompt 32: Write a 90-day “hate-to-content” retrospective
Purpose: Once a quarter, convert your hate-comment data into a strategic plan.
ROLE: You are a content strategist trained in JTBD, StoryBrand,
and the Pole Position framework by Sam Parr.
CONTEXT: Over the last 90 days, I have collected 200+ comments
across [PLATFORMS]. About 60 were negative or skeptical. I
want to convert the data into a 90-day retrospective and a
plan for the next quarter.
DATA: [PASTE_COMMENT_DUMP_OR_SUMMARY]
TASK:
1. Summarize the top 5 audience concerns in priority order.
2. For each concern, list:
- 1 piece of content I made that addressed it well
- 1 piece of content I made that did not
- 1 content idea I should make next quarter
3. Identify 3 new content pillars I should consider testing.
4. Identify 1 content pillar I should retire.
5. Suggest 1 community policy change.
6. Recommend the single biggest bet for the next 90 days, with
a one-paragraph justification.
CONSTRAINTS:
- Base all recommendations on the data, not vibes.
- Do not recommend "post more" unless the data supports it.
OUTPUT FORMAT:
Top 5 concerns (priority order): ...
Per-concern analysis: ...
New pillars to test: ...
Pillar to retire: ...
Community policy change: ...
Biggest bet for next quarter: ...Pro tip: Run this retrospective once a quarter with your team. It is the highest-leverage 90 minutes in your content calendar.
Comparison TABLE: prompt category vs. comment type vs. output
This table is the cheat sheet. Screenshot it. Pin it to your Notion. Use it every Friday.
| Prompt category | Comment type it fits | Best output format | When to run it | Primary KPI it moves |
|---|---|---|---|---|
| Comment-mining (1-5) | Lots of comments, no time | Notion database, content calendar | Weekly | Time saved, content backlog size |
| Reply & moderation (6-11) | Single hot comment or thread | 1-3 sentence reply, DM, or moderation action | Daily, in the moment | Reply speed, community sentiment |
| Reframe & rebuild (12-17) | A single comment with a real concern | StoryBrand paragraph, hook bank, NVC reply | 2-3 times a week | Comment quality, save rate |
| New content from comments (18-23) | A comment you want to feature in a post | Carousel, TikTok, Substack, podcast outline | Weekly | Reach, watch time, follower growth |
| Multi-platform post (24-28) | A long-form piece you want to spread | 7-day, 7-platform plan, Stories, thread | Weekly | Cross-platform engagement |
| Crisis & brand-safety (29-32) | Coordinated attack, doxx, threat | Holding statement, moderation plan, retrospective | As needed, plus quarterly drill | Brand trust, retention |
The reason this table matters: most creators run prompts 1-5 all the time and prompts 29-32 never. The prompts you do not run are usually the ones that would save you the most pain.
People Also Ask: 10 questions about ChatGPT prompts for turning hate comments into content
These are the questions real creators type into Google. Each answer is short, direct, and AEO-snippet-friendly.
1. Do hate comments actually increase engagement? Yes. Posts that earn mid-length comments (50-99 characters) see 151.6% higher engagement on average, per Hootsuite’s 2026 Social Trends report. Hate comments are a subset of comments. They count.
2. Should I delete hate comments or reply to them? Delete slurs, spam, and threats. Reply to thoughtful criticism. Ignore drive-by snark. The 80/20 rule applies: most “hate” is actually VIBE, not CRISIS. Use the comment triage framework in Prompt 8.
3. Can ChatGPT write replies to hate comments for me? Yes, but only if you give it the framework. Paste the comment into Prompt 6 or Prompt 9. Always read the reply before posting. Never post a reply you would not say out loud.
4. How do I respond to a hate comment without sounding defensive? Use Marshall Rosenberg’s Nonviolent Communication: observation, feeling, need, request. Use Prompt 16. Keep it under 100 words. No “but.” No “should.”
5. How do I know if a hate comment is a real concern or just trolling? Run Prompt 15. It separates “real person, real concern” from “trolling bad-faith actor.” Default to “real concern” until the language signals otherwise.
6. What is the best platform to repurpose a hate comment into content? TikTok for reach, LinkedIn for trust, Substack for depth. The same comment can fuel all three. Use Prompt 24 to plan a 7-platform rollout.
7. How often should I turn hate comments into content? Once a week is a healthy cadence. More than that and you start feeding the trolls. Less than that and you are leaving the best research on the table.
8. Can I quote the hater by name? No. Paraphrase. If you must quote, get their consent via DM. Public callouts almost always backfire. Use Prompt 28 for the consent flow.
9. How do I prevent hate comments from hurting my mental health? Use a 30-minute daily window for moderation. Use a tool like Agorapulse or Hootsuite to batch replies. Use the triage prompt to label comments as ENGAGE, ARCHIVE, or BLOCK. Outside the window, close the app.
10. Will ChatGPT-generated replies sound like AI? Only if you let them. The prompts above force specific frameworks (Crucial Conversations, NVC, StoryBrand). Edit the output to add your voice. Aim for 70% AI draft, 30% you. That blend reads human and saves hours.
A 14-day “hate-to-content” sprint
This is the plan I would run if I were starting from scratch tomorrow. Fourteen days. No fancy tooling required beyond ChatGPT, Notion, and a scheduler like Buffer.
Days 1-2: Audit and triage
- Export your last 100 comments from each platform into a single Notion database.
- Run Prompt 8 on each batch. Label every comment as ENGAGE, ARCHIVE, or BLOCK.
- Save the ENGAGE comments as a new database called “Hate-to-content raw.”
Days 3-4: Mine and cluster
- Run Prompt 1 on the entire “raw” database.
- Save the top 5 concerns as a new database called “Concerns.”
- For each concern, link the original comments.
Days 5-7: Mirror and reply
- Spend 30 minutes a day replying using Prompt 6 or Prompt 16.
- Aim for 3 thoughtful replies per day.
- Pin the best reply in your Notion “Mirror Bank.”
Days 8-10: Make content
- Pick the top 2 concerns from “Concerns.”
- For each, produce one piece of content using Prompt 18 (carousel), Prompt 19 (TikTok), or Prompt 20 (Substack).
- Post one piece per day for 3 days.
Days 11-12: Multiply
- Take the best-performing piece from days 8-10 and run Prompt 24 to repurpose it across 7 platforms.
- Schedule with Buffer or Hootsuite.
Day 13: Crisis drill
- Fill in the one-page crisis template from Prompt 30.
- DM your moderators the doc.
Day 14: Retrospective
- Run Prompt 32 on the last 14 days of comments.
- Decide your top bet for the next 90 days.
That is it. Fourteen days. By the end, you will have a content calendar, a reply bank, a crisis plan, and a clearer view of your audience than you have had in years.
Common mistakes to avoid
I have watched a lot of creators try this. The same five mistakes kill the results every time.
- Mistake 1: Replying in the same emotional register as the hater. If they are angry, you are calm. If they are sarcastic, you are direct. Prompt 9 is your safety net.
- Mistake 2: Posting the reply without editing. AI outputs read AI. Add 30% of your voice before you post. The audience can tell.
- Mistake 3: Quoting the hater by name. Even with a “but” attached. Use Prompt 28 to feature the idea, not the person.
- Mistake 4: Treating every hate comment as a content opportunity. Some comments are just abuse. Use Prompt 29 to triage. Not every comment deserves a post.
- Mistake 5: Skipping the framework. A prompt without the Mine, Mirror, Make, Multiply structure will produce generic output. Always start with the framework, then the prompt.
Final word
Your haters are not a problem to solve. They are an audience to serve. They are handing you a free focus group, a free A/B test, and a free content brief, all in a single message. Most creators will delete it. You will not.
Use the 32 prompts. Run the 14-day sprint. Post the carousel. Film the TikTok. Write the Substack. Reply to the comment with a mirror, not a match.
The work is not in the prompt. The work is in the reply, the post, and the next audience member who feels seen because you answered the question they were too shy to type.
That is the whole game.