_edited.png){kind=small}
The 20 AI Marketing Prompts That Replaced My $100K Marketing Team
- Ben Angel
- 14 minutes ago
- 37 min read
A full-stack marketing team — strategist, copywriter, SEO lead, ads buyer, analyst — runs about $100,000 a year, minimum. I know because I priced it. Twice.
The reason AI marketing prompts have become one of the most valuable assets in business is simple: in the first week of June 2026, the Big Three made hiring that team almost irrational.
On June 2, OpenAI shipped Codex role-specific plugins — bundles of marketing, analyst, and design workflows that previously needed an engineer or a stack of SaaS. Two days later, Google announced Gemini Spark, a 24/7 AI marketing agent that briefs creative, generates assets, and optimizes ad campaigns on its own. Three days after that, Anthropic pushed Claude into Level 4 autonomy with native marketing slash commands like /campaign-plan and /seo-audit — replacing prompt engineering with enterprise templates.
One week. Three giants. One quiet announcement: the marketing team you were about to hire just got automated.
This is what The Wolf Is at The Door warned about — the giants don't compete for your attention anymore. They compete for your job functions. And solopreneurs and small business owners who don't learn to wield these agents are about to get out-shipped by 17-year-olds with $20 Claude subscriptions.
So here are the 20 AI marketing prompts I now use to do the work of that $100K team. Every one has been pressure-tested in ChatGPT, Claude, or Perplexity Computer. None of them are the recycled "fill-in-the-blank" prompts you've seen 400 times. These are the ones the top 1% actually run.
Why Marketing Prompt Lists Are Making You Worse
I've read the lists. The ones with 500, 1,000, even 2,000 "AI marketing prompts." Most of them look like this:
"Create a social media campaign for <business> targeting <persona> by showcasing our <USP> and addressing their <challenges>."
That's not a prompt. That's a fill-in-the-blank template. You paste in your business, you get back generic AI slop, you wonder why your engagement is flat. The 1,000-prompt PDF wasn't designed to make you better at marketing. It was designed to make a creator look productive.
Here's the uncomfortable truth: the bottom 50% of marketers ask AI to do tasks. The top 1% architect prompts that produce strategic decisions.
The gap isn't quantity. It's structure. And once you see the structure, you can never un-see it.
In This Article
The Setup
Category 1: Content That Compounds
Category 2: Audience & Authority
Category 3: Email & Offers
Category 4: Research & Decisions
Category 5: Business Operations
The Close
The Top 1% Upgrade Framework
Every prompt below was rebuilt using this 8-dimension framework. Use it to audit any prompt you've ever copy-pasted from the internet:
1. Role assignment: Bottom 50% way: No role → Top 1% way: Names a specific expert archetype + the principle they're known for
2. Context injection: Bottom 50% way: Empty <business> placeholder → Top 1% way: Forces you to paste real assets — emails, transcripts, reviews, analytics
3. Reasoning method: Bottom 50% way: None → Top 1% way: Embeds a thinking framework — chain-of-thought, steelman, pre-mortem
4. Output structure: Bottom 50% way: Free-form essay → Top 1% way: A defined schema — table, ranked list, decision matrix
5. Self-critique loop: Bottom 50% way: None → Top 1% way: "Now critique your own answer and give me a sharper v2"
6. Anti-slop guardrails: Bottom 50% way: None → Top 1% way: Explicit banned phrases, banned clichés, banned LLM tells
7. Chained workflow: Bottom 50% way: One-shot → Top 1% way: Prompt 1's output becomes Prompt 2's input
8. Tool leverage: Bottom 50% way: None → Top 1% way: Tells the model to pull from web, files, connectors, live data
If a prompt doesn't have at least 5 of these, it's not a top 1% prompt. It's a fill-in-the-blank. Save your time.
Now — the 20.
CATEGORY 1: CONTENT THAT COMPOUNDS
Content is the only marketing activity that keeps paying you while you sleep. A great YouTube video filmed in 2024 is still generating leads in 2026. A great blog post is still ranking three years later. The reason most solopreneurs underperform here isn't laziness — it's that they treat each piece as a one-off instead of a compounding asset.
These five prompts treat every piece of content as a system. They reverse-engineer what worked, multiply it, and squeeze 10x the leverage out of every minute you spend filming or writing. (New to ChatGPT? Start here first.)
From The Wolf Is at The Door: "In the age of disruption, the irreplaceable creator isn't the one who works harder. It's the one who makes every minute of work do the job of ten."
Prompt #1: The Hook Forensics Engine
Best in: Claude (long context handles 30+ transcripts at once)
Most creators rewrite hooks based on vibes. The top 1% do hook forensics — they extract the exact linguistic and structural patterns from every video that worked, then weaponize those patterns on the next one. This prompt does in 90 seconds what a hook coach charges $5,000/month for.
Pro tip: Run this prompt inside Perplexity Computer and connect your YouTube Data — it'll pull view-velocity stats automatically and skip the manual paste.
THE PROMPT:
You are a hook strategist trained in the direct-response tradition of David Ogilvy (the "5x more people read the headline than the body" rule) and the modern attention engineering of Mr. Beast (front-loaded curiosity + visual escalation). I will paste my last 20 video titles + first 7 seconds of script + their 24-hour view count. Your job, in this exact order: 1. Rank all 20 from highest to lowest view velocity. 2. Identify the 3 structural patterns my top 5 hooks share that my bottom 5 don't — be specific about sentence structure, the verb in the first 3 words, and the emotional trigger. 3. Identify the 3 patterns my bottom 5 share that quietly killed them. 4. Build me a 10-hook template library based on the winning patterns, each labeled with the trigger it activates (curiosity gap, contrarian claim, named enemy, specificity, pattern interrupt, etc.). 5. For my next video on [INSERT TOPIC], generate 7 hook options using 7 different winning patterns from my own data. 6. Critique your own 7 hooks. Identify the 2 weakest. Rewrite them. Output as a single table with columns: Pattern Name | Hook | Trigger Activated | Why It Works Against My Data. Banned: "In a world where," "What if I told you," "You won't believe," "Here's the thing," any sentence starting with "So,". Do not invent hooks not grounded in my data.
What you get: A hook template library trained on YOUR audience, not someone else's.
What to do next: Run this monthly. Your hook bank gets sharper with every cycle.
Prompt #2: The Video-to-Empire Repurposing Architect
Best in: ChatGPT (best multi-format generation in one pass)
One video script can become 14 pieces of content across 6 platforms — if you architect the repurposing instead of guessing it. Most creators get 3-4 pieces because they're freestyling. This prompt squeezes every drop of leverage from a single 10-minute filming session.
Pro tip: Use this prompt in Perplexity Computer so it can save each output directly to your connected Google Drive folder for your editor. It leverages 19+ AI models, if it gets stuck, it spins up a new agent to solve the problem.
THE PROMPT:
You are a content repurposing strategist trained in Gary Vaynerchuk's content pyramid model (one pillar piece → micro-content across every channel) and Justin Welsh's solopreneur leverage system (every piece must work without context). I'll paste my full video script below. Build me a 14-asset repurposing plan with one asset per row, in this exact priority order: 1. 3 LinkedIn long-form posts — each pulling a different angle (story, framework, contrarian take). Each must hook in line 1, deliver in lines 2-7, end with a saveable insight. 2. 2 X/Twitter threads — one teaching, one storytelling. 8-12 tweets each. 3. 1 Instagram carousel — 8 slides, hook on slide 1, payoff on slide 7, CTA on slide 8. 4. 3 YouTube Shorts/Reels scripts — different hooks, different angles, each under 45 seconds. 5. 1 email for my list — driving them to the full video, written in conversational voice. 6. 1 blog post outline — H1, 5 H2s, target keyword, internal-link suggestions based on the script's themes. 7. 3 Instagram/LinkedIn graphic quote candidates — pulled verbatim from the script, ranked by shareability. For each asset, include: Platform | Hook (first line) | Body angle | CTA | Native format note (e.g. "LinkedIn algorithm favors line breaks every 2 sentences"). Self-critique: identify the 2 assets least likely to perform. Rewrite them. Banned: "Game-changer," "unlock," "in today's fast-paced world," any sentence that could have been written about any other creator.
What you get: A full editorial calendar from one video.
What to do next: Hand the table to your editor or VA. You just compressed a week's worth of content into one filming day.
Prompt #3: The Blog Post That Ranks AND Converts
Best in: Perplexity Computer (live SERP analysis + Search Console pull in one shot)
99% of blog post prompts produce SEO content that ranks but doesn't convert, OR conversion content that converts but doesn't rank. The top 1% prompt does both in one pass — because it forces the AI to analyze the actual ranking SERP first, then write to outperform it.
Pro tip: This is exactly the kind of multi-step research-and-write workflow Perplexity Computer was built for — it can pull live SERPs, your competitor's actual page, and your Search Console data in a single prompt.
THE PROMPT:
You are a content strategist trained in the SEO frameworks of Eli Schwartz (product-led SEO that converts) and Rand Fishkin (search intent as the foundation of every piece). I want to write a blog post targeting the keyword [INSERT KEYWORD]. Step 1: Pull the top 5 ranking URLs for that keyword right now. For each, identify: word count, headers used, the angle they took, what they're missing, and the conversion mechanism (if any). Step 2: Identify the "content gap" — the angle no one in the top 5 is taking that would make my post 10x more useful. Step 3: Identify the "conversion gap" — the offer, lead magnet, or CTA placement none of them are using. Step 4: Build me a full blog post outline that: Targets the primary keyword in the H1, first 100 words, and 2 H2s; Includes 5 H2s and 8 H3s mapped to long-tail variations; Specifies the 3 internal links I should weave in (you'll need to ask me for my site's existing relevant content); Specifies a meta description under 155 chars with the keyword in the first half; Places exactly 2 CTAs: one mid-post (lead magnet), one closing (offer). Step 5: Write the opening 200 words in my voice. Pattern-interrupt hook → tension → promise. Step 6: Score the outline 1-10 against the top-ranking URL on three dimensions (depth, uniqueness, conversion design). If any score is under 8, rewrite that section. Banned: Filler phrases, "in this article we'll cover," "let's dive in," any throat-clearing intro.
What you get: A blog post architected to outrank AND outconvert the current #1.
What to do next: Write to the outline. Publish. Track ranking weekly.
Prompt #4: The Video Script That Holds Retention Past 70%
Best in: Claude (handles 20+ retention graphs in context without losing the pattern)
The reason most "AI video script" prompts produce garbage is that they optimize for what the script says, not for what makes a viewer stay. The top 1% prompt forces the AI to reverse-engineer your actual retention drops first — then writes against your specific viewer-loss patterns.
Pro tip: Paste your last 10 YouTube analytics retention curves as screenshots. Claude can read them and find the exact second your viewers leave.
THE PROMPT:
You are a retention engineer trained in the storytelling architecture of Robert McKee (every scene must escalate stakes or it gets cut) and the platform mechanics of MrBeast (pattern interrupts every 30-60 seconds to break the brain's prediction loop). I'll paste/upload the retention curves from my last 10 videos + their scripts. Step 1: Identify the average second-mark where my viewers drop off in the first 30 seconds. Tell me what's structurally wrong with my opens. Step 2: Identify my 3 most common mid-video drop points (typically minutes 2-4). Tell me what content choice triggered each one. Step 3: Build a 10-minute video script for [INSERT TOPIC] engineered specifically to defeat my drop points, using: A 0-3 second hook that addresses my open-rate weakness; A 7-second "promise + stakes" frame; Pattern interrupts at exactly the seconds my audience historically leaves; A "loop open" in the first minute that pays off in the last minute; One "story spike" every 90 seconds (specific story, named person, exact dollar figure, or dated event — never abstract); A 30-second closing CTA that frames the next action as inevitable, not optional. Output the script with timestamps and one annotation per section: "this beat targets [my specific weakness X]." Self-critique: identify the section most likely to lose me a viewer based on my history. Rewrite it. Banned: "Hey guys," "in this video we're going to," "make sure to like and subscribe," "without further ado." Open with the hook. Period.
What you get: A video script engineered against YOUR retention weaknesses, not generic ones.
What to do next: Film it. Compare the retention curve. Iterate the next script using the new data.
Prompt #5: The 30-Day Content Engine From One Audit
Best in: Perplexity Computer (web research + Search Console + your historical content in one workflow)
Most creators sit down Monday morning and ask "what should I post this week?" The top 1% never asks that question — because they've already engineered a 30-day content calendar from a single workflow that mines what their audience actually wants. This prompt is the workflow.
Pro tip: Run this in Perplexity Computer so it can connect your Search Console, your past content library, and live web research in one chained workflow — this is the kind of multi-tool agentic work it was built for.
THE PROMPT:
You are a content strategist trained in the audience-research methodology of Amanda Natividad (audience-first SEO via real community signals) and the editorial systems of Anne Handley (every piece must earn its slot on the calendar). Build me a 30-day content calendar using this exact workflow: Phase 1 — Mine my audience (do this first): Pull the top 20 questions being asked in [my niche] across Reddit, X, and Quora in the last 30 days; Pull the 10 highest-impression, lowest-CTR queries from my Google Search Console (those are unmet demand I'm being shown for but failing); Pull the comments from my last 5 YouTube videos and extract the 10 most frequently repeated questions/objections. Phase 2 — Find the gaps: For each of the 40 demand signals you found, check if I already have content addressing it; Identify the 12 strongest gaps where demand is high but I have nothing. Phase 3 — Build the calendar: 30-day calendar, 1 piece per weekday (20 total); Mix: 12 short-form videos, 4 long-form videos, 2 blog posts, 2 carousels; Each entry: Date | Format | Topic | Hook | Primary Platform | Demand Signal It Addresses | Internal Link Target | Lead Magnet CTA; Cluster pieces into 4 weekly narrative arcs so each week tells a story, not just lists topics. Phase 4 — Stress test: Identify the 3 weakest entries on the calendar. Replace them with stronger angles; Identify the 1 piece most likely to go viral. Tell me why and how to film it. Banned: Generic topics ("AI tips," "productivity hacks"). Every topic must be specific enough that it could only have been written this month, in this market.
What you get: 30 days of content, every piece tied to real audience demand and a strategic outcome.
What to do next: Block one filming day per week. Execute the calendar. Re-run this prompt monthly.
CATEGORY 2: AUDIENCE & AUTHORITY
Content gets discovered. Audience gets built. The difference is whether the people consuming your work end up in a list you own — your email subscribers, your followers who actually open notifications, the small army that ships your launches. The bottom 50% of creators post and hope. The top 1% architect every piece to convert strangers into followers, followers into subscribers, and subscribers into buyers.
These four prompts treat audience-building as a forensic discipline, not a vibe. (This is one of the core themes in my breakdown of how AI is reshaping solopreneur business.)
From The Wolf Is at The Door: "Followers are rented. Email subscribers are owned. The irreplaceable creator builds owned audience first, and treats every platform as a feeder."
Prompt #6: The LinkedIn Long-Form Authority Builder
Best in: Claude (best at long-form structural writing + voice match)
LinkedIn is the most underused growth channel for solopreneurs because most creators write LinkedIn posts like blog posts (boring) or like Twitter posts (too short to land). The top 1% write to a specific 8-beat structure that the LinkedIn algorithm and human readers both reward. This prompt builds to that structure every time.
Pro tip: Run this in Perplexity Computer — it can read your last 30 LinkedIn posts AND their engagement data to match your highest-performing voice patterns.
THE PROMPT:
You are a LinkedIn long-form strategist trained in Justin Welsh's solopreneur LinkedIn frameworks (every post must be readable in 9 seconds OR pull the reader into a 90-second story) and Lara Acosta's authority-narrative model (story + framework + identity reframe). I'll paste my last 10 LinkedIn posts with view + reaction + comment counts. Step 1: Identify what my top 3 posts have in common — sentence length, opening pattern, body structure, hook type, ending type. Be specific. Step 2: Identify what my bottom 3 share that quietly killed them. Step 3: Build a LinkedIn post on [INSERT TOPIC] using my winning patterns and this exact 8-beat structure: 1. Hook line — 5-9 words, pattern-interrupt or contrarian claim; 2. Tension — a single sentence that creates a "wait, what?" gap; 3. Story or setup — 2-4 short lines, concrete people/numbers/dates; 4. The shift — the moment of insight or reversal; 5. The framework or lesson — 3-5 bullet points or numbered beats; 6. The reframe — 1 sentence that changes how the reader sees themselves; 7. The save-worthy insight — 1 quotable line built for screenshots; 8. The soft CTA — a question that invites comments, never a "buy now". Format: every 1-2 sentences gets its own line break. Total length 180-250 words. Self-critique: identify which of the 8 beats is weakest. Rewrite the entire post fixing that beat. Banned: "I'm excited to share," "thrilled to announce," "humbled," "blessed," "let that sink in," "the best part?," "here's the kicker," any post that could have been written by any other LinkedIn creator.
What you get: A LinkedIn post architected to your audience's documented preferences.
What to do next: Post at 7am or 12pm local. Reply to every comment in the first hour.
Prompt #7: The Instagram Carousel That Stops the Thumb
Best in: ChatGPT (best balance of copy + structural layout reasoning)
Instagram carousels are the highest-saving-rate format on the platform — but only if slide 1 stops the thumb AND slides 2-9 escalate the curiosity. Most creators write slide 1 like a headline and the rest like an essay. The top 1% engineer each slide as a discrete cliffhanger.
THE PROMPT:
You are an Instagram carousel designer trained in Jenny Hoyos's short-form attention engineering (every beat must escalate or it gets cut) and Daniel Pink's "When"-style storytelling (build → twist → resolution). I'll paste the topic + my brand voice description + a link to my last 5 best-performing carousels. Step 1: Identify the structural pattern across my top 5 — slide 1 hook type, slide 2-3 setup, where the twist lands, how the CTA on the final slide is framed. Step 2: Build an 8-slide carousel on [INSERT TOPIC] using this exact beat structure: Slide 1 (Hook): 5-8 words maximum, must trigger a curiosity gap or contrarian claim. NEVER use a question as the hook. Slide 2 (Reframe the stakes): "Here's what most people miss..." Slide 3-5 (Build): Three escalating insights, frameworks, or steps. Each one must be screenshot-worthy on its own. Slide 6 (Twist): The non-obvious insight that recontextualizes everything above. Slide 7 (Payoff): The "so what you should do is..." — concrete, actionable, specific. Slide 8 (CTA): Soft offer — lead magnet or follow prompt. Never a sales ask. For each slide, output: Slide # | Headline | Body copy (under 25 words) | Visual direction note (e.g. "1 bold word centered, dark background"). Then generate the caption: hook line + 3-5 short paragraphs expanding the carousel + CTA + 8 hashtags (mix of niche-specific and broader). Self-critique: rate each slide 1-10 on thumb-stop potential. Rewrite any slide under 8. Banned: Slide 1 starting with "How to," "5 ways," "The ultimate guide." Captions ending with generic engagement-bait questions.
What you get: A carousel structured to maximize saves and shares — Instagram's two best signals.
What to do next: Send to Canva, ship within 24 hours. Carousel relevance decays fast.
Prompt #8: The Viral Hook Engineering Lab
Best in: Claude (handles batch generation of 30+ variations with consistent quality)
Every viral creator has one secret: they don't write a hook. They write 30, and then they pick the 1 that's mathematically structured to win. This prompt is the lab. Use it before you write anything else — script, post, email, headline.
Pro tip: Save the output to a "hook bank" doc in Google Drive. Pull from it for every new piece of content for the next 90 days.
THE PROMPT:
You are a hook engineer trained in Alex Hormozi's hook physics model (hooks must do one of: shock, intrigue, identify-with) and Roy Williams's Wizard of Ads dramatic-element framework (specificity + sensory detail + emotional contrast). I want to write a piece of content on [INSERT TOPIC] for [INSERT PLATFORM: YouTube/LinkedIn/X/Email]. Generate exactly 30 hooks, organized into 10 categories of 3 hooks each: 1. Curiosity gap — withhold the answer, force the click; 2. Contrarian claim — challenge a widely held belief; 3. Named enemy — call out a specific villain (platform, practice, person, archetype); 4. Specific number — exact dollar, exact time, exact result; 5. Pattern interrupt — break the reader's prediction; 6. Status threat — the reader's identity is at stake; 7. Inside knowledge — "what nobody tells you about X"; 8. Personal stakes — "I lost $X by doing Y"; 9. Time-bomb urgency — a deadline or shrinking window; 10. Identity reframe — "stop being X, start being Y". For each hook, output: Hook | Category | Predicted trigger (curiosity/anger/fear/aspiration/identity) | Best-fit platform. Now rank all 30 from highest to lowest projected performance based on the platform I named. Tell me the top 5 and why. Self-critique: identify the 3 worst hooks. Rewrite them to fix the specific failure (vague, cliché, weak verb, no stakes). Banned: Questions as hooks ("Did you know...?"), "In a world where," "Here's the thing about," anything starting with "So,". Every hook must be under 12 words.
What you get: 30 hooks, ranked. You pick the top 1-3 and write to them.
What to do next: Use this every Monday before the week's content. It's a 10-minute investment that 2x's everything you ship.
Prompt #9: The Lead Magnet That Earns the Email
Best in: Perplexity Computer (web research on competitor lead magnets + your offer alignment in one pass)
Most lead magnets are PDFs nobody reads, which is why most email lists grow slowly. The top 1% build lead magnets that are so specific, so painfully useful, that people would pay for them. This prompt designs that lead magnet — and writes the whole thing.
Pro tip: In Perplexity Computer, this prompt can also research your top 5 competitors' lead magnets in the same session — so you build something demonstrably better than what's already converting.
THE PROMPT:
You are a lead magnet strategist trained in Sahil Bloom's "specific problem → specific tool" lead magnet philosophy (the best lead magnets solve one painful problem in under 20 minutes of consumption) and Joanna Wiebe's voice-of-customer methodology (mine real customer language to position the magnet). I'll paste: (1) my offer description, (2) my ICP description, (3) 10 verbatim DMs/emails from my audience describing their problems. Phase 1 — Mine the data: Extract the top 5 pain points using only language my audience actually used; Identify the single most painful, specific, unsolved problem. Phase 2 — Design the lead magnet: Lead magnet title (under 10 words, must contain a number or specificity hook); Format (checklist, template, swipe file, mini-course, audit tool — recommend the format that fits the problem best); The "20-minute promise" — what the reader can do after consuming it; Why this magnet is a perfect bridge to my main offer (logical next step, not random). Phase 3 — Build the magnet: Full table of contents with 5-7 sections; Write the first section in full (300-500 words); Outline the remaining sections with one-line descriptions; Design the "results page" — what the reader sees after they finish it. Phase 4 — Build the opt-in page: Headline (under 12 words, benefit-driven, includes the specific outcome); Subhead (1-2 sentences positioning who it's for and who it's NOT for); 3-5 bullet points of what's inside (each bullet leads with the outcome, not the deliverable); Button copy options x3 (action verb + benefit). Self-critique: would a stranger in my niche actually opt in? Identify the weakest element. Rewrite it. Banned: "Ultimate guide," "everything you need to know," generic numbers like "10 tips," anything that doesn't survive the "would they pay $50 for this?" test.
What you get: A lead magnet designed to convert cold traffic and bridge to your offer.
What to do next: Ship the first section. Use it for 2 weeks. Track opt-in rate. Iterate.
CATEGORY 3: EMAIL & OFFERS
Email is where the money is made. Always has been. Every other channel is rented attention — email is owned. But most solopreneurs have an email list that "exists" without actually converting because they treat each email as a standalone task instead of as one move in a chess game.
These four prompts treat your email list as a revenue system. They diagnose what's broken, optimize what's leaking money, and architect launches that actually convert. (The AI Mastery system I broke down here goes deeper into the email-to-offer connection.)
From The Wolf Is at The Door: "Your email list is the only marketing asset Mark Zuckerberg can't take from you in a single platform decision. Treat it accordingly."
Prompt #10: The Email Campaign Autopsy
Best in: Claude (handles full email + analytics paste-in, finds patterns across many sends)
When an email tanks, most creators shrug and write the next one. The top 1% perform an autopsy — they identify the exact mechanical failure point and fix it forever. This prompt is the autopsy tool. Run it after every send.
THE PROMPT:
You are an email diagnostician trained in the direct-response analysis tradition of John Caples (every element of the email is testable and falsifiable) and the lifecycle email methodology of Val Geisler (every email is one move in a longer customer story). I'll paste: the full subject line, preview text, full email body, send time, open rate, click rate, conversion rate, and unsubscribe rate. Run this exact 6-point diagnostic: 1. Subject line autopsy: Score 1-10 on curiosity, specificity, threat-or-reward signal, sender-relevance, character count. Identify the single weakest dimension. 2. Preview text autopsy: Did it complete or repeat the subject? It should complete. If it repeated, flag it. 3. Opening line autopsy: Did line 1 either pull the reader into a story or pay off the subject's promise? If neither, flag the failure type. 4. Body autopsy: Identify the exact paragraph where reader attention most likely dropped. Tell me why structurally (too long, too abstract, story shifted, value not yet delivered). 5. CTA autopsy: Was there exactly one CTA? Was it specific? Did the surrounding copy frame the click as the obvious next step or as an imposition? 6. Send-time autopsy: Was the send time aligned with my list's historical open behavior? If not, flag it. Then output: The single root cause of the underperformance (one sentence); 3 specific fixes for the next send (each fix tied to a specific autopsy finding); A rewritten subject line + opening paragraph using my voice. Self-critique: am I being too harsh or too generous? If my fixes are vague, rewrite them with more specificity. Banned: Generic advice like "make it more engaging," "shorten it," "test more subject lines." Every recommendation must be specific enough to execute in 10 minutes.
What you get: A surgical breakdown of why an email underperformed and exactly what to fix.
What to do next: Apply the 3 fixes to your next send. Rerun the autopsy. Compounding insight.
Prompt #11: The Sales Page Reverse-Engineer
Best in: Perplexity Computer (live URL fetch of competitor sales pages + analysis)
You don't need to write a sales page from scratch. You need to reverse-engineer the 3 sales pages currently outselling yours and steal their structural moves — without copying their copy. This prompt does the forensics.
Pro tip: Run this in Perplexity Computer — it can fetch and analyze 3 live competitor sales pages in the same session you write yours.
THE PROMPT:
You are a sales page strategist trained in April Dunford's positioning methodology (the customer's alternative is always "do nothing" or "use a substitute" — your page must defeat both) and Joanna Wiebe's pain-dream-fix conversion copy framework. I'll give you 3 competitor sales page URLs in my niche + my offer description. Phase 1 — Reverse-engineer the 3 competitor pages: For each, map the page architecture beat-by-beat (hero, problem agitation, solution intro, mechanism, social proof, offer stack, objections, CTAs, guarantee); Identify the dominant emotional appeal of each (fear, status, transformation, time, identity); Identify each page's "alternative-killing move" — how it argues against doing nothing and against substitutes; Identify the structural gap none of the 3 are filling — the angle no one is using. Phase 2 — Architect my page: Build my page outline beat-by-beat, using the structural moves that win across all 3 competitors AND filling the gap they're missing; Specify the dominant emotional appeal that differentiates my page from all 3; Specify exactly where I'll plant the "alternative-killing move"; Specify the hero headline + subhead + first CTA copy; Specify the offer stack reveal moment and the 5 components I'll stack; Specify the 3 objections to address and where each addresses lands; Specify the guarantee position and copy. Phase 3 — Write the hero section in full: Headline (under 12 words, addresses the pain at the level my customer feels it); Subhead (1-2 sentences, who it's for + the outcome); First CTA button copy; Sub-CTA microcopy (e.g. "instant access — no card required to start"). Self-critique: score my outline against the strongest competitor on persuasion architecture 1-10. If under 9, identify the weakest beat and rewrite it. Banned: Generic claims, "transform your business," "unlock your potential," "in just 30 days," any phrase that could be on any other sales page.
What you get: A sales page architected to beat the current top 3 in your market.
What to do next: Build the page. Ship a small ad budget at it. Measure conversion against the competitor baseline.
Prompt #12: The Checkout & Order Form Conversion Lift
Best in: Claude (best at granular UX critique with structured output)
You're losing money at the checkout step and you don't know it. Tiny order form decisions — field order, button copy, trust signals, urgency placement — compound into double-digit conversion swings. Most solopreneurs never audit their checkout. The top 1% audit it every 90 days.
THE PROMPT:
You are a checkout conversion specialist trained in Peep Laja's CXL conversion research methodology (every checkout element is testable and most decrease conversion when added) and the friction-reduction principles of Luke Wroblewski (each form field cuts conversion by ~5-7% on average — earn every field). I'll paste/screenshot my current order form + checkout flow + my last 30 days of cart abandonment rate. Run this exact audit: 1. Field-by-field analysis: For every field on my form, decide: keep, remove, or make optional. Justify each call against the 5-7% per-field rule. 2. Trust signal audit: What trust elements are present? Money-back guarantee, secure checkout badges, testimonial near the button, named-customer logos? What's missing and where exactly should it be placed? 3. Button copy audit: Is the button copy action + benefit? Score 1-10. If under 9, write 3 alternatives. 4. Urgency / scarcity audit: Is there a legitimate reason to act now? If yes, is it visible without scrolling? If no, recommend whether to add one (and only if it's real — never fake scarcity). 5. Price-presentation audit: How is the price framed? Is the value-stack visible above the price? Is the comparison (monthly vs annual, "vs hiring an X") clearly shown? 6. Post-click experience audit: What happens immediately after the button is clicked? Is the loading state confidence-building or anxiety-inducing? 7. Mobile audit: Are buttons thumb-reachable? Is autofill enabled? Is the form scrollable without zoom? Then output: Top 3 changes ranked by projected conversion lift (each with the specific change and why it will move the needle); The single change to ship today (lowest effort, highest expected lift); The 1 thing to A/B test next week (after shipping the top 3). Self-critique: am I recommending changes based on best practice or based on this specific form? If too generic, rewrite recommendations to be specific to my form. Banned: Generic CRO advice. Every recommendation must reference a specific element on my form.
What you get: A prioritized checkout improvement roadmap with projected impact.
What to do next: Ship the top change today. Test the next change in 7 days. Compound the lift.
Prompt #13: The Launch Email Sequence That Doesn't Burn the List
Best in: Claude (long context handles a full 7-email sequence with continuity check)
Most launch sequences treat the list like a vending machine — pump out 5 emails, take the money, leave a trail of unsubscribes. The top 1% architect launches where every email is valuable enough to forward, even if the reader doesn't buy. This prompt builds that sequence.
THE PROMPT:
You are a launch strategist trained in Jeff Walker's Product Launch Formula (pre-launch content builds belief before pitch) and the relationship-first launch model of Amy Porterfield (every email earns the right to the next one through value, not pressure). I'll paste: my offer description, my ICP, my past launch performance data if I have any, my brand voice description. Build me a 7-email launch sequence over 10 days, structured as: Email 1 (Day 1): Story-driven opener. Reveal a transformation, not the offer. End with a soft promise of what's coming. No pitch. Email 2 (Day 3): Reframe the problem in a way the reader hasn't considered. Plant the seed for why the standard solution fails. Email 3 (Day 5): Introduce the offer with the "mechanism" — why it works, not what it is. Soft CTA to the sales page. Email 4 (Day 6): Address the #1 objection with a story (not a list of features). Email 5 (Day 8): Social proof email — one specific transformation story with concrete numbers. Email 6 (Day 9): Address the #2 objection with a reframe. Mid-strength CTA. Email 7 (Day 10): Last-call urgency email — must be honest about the deadline. Strong CTA. For each email, output: Subject line (with 3 alternatives) | Preview text | Hook line | Body angle | Specific story or proof point | CTA copy | The single "earn the next open" insight (the saveable line). Continuity check: does the narrative escalate across the 7? Is each email valuable on its own? Identify the email that's most likely to trigger an unsubscribe — rewrite it. Self-critique: which email feels most pressure-driven instead of value-driven? Rewrite it to lead with value. Banned: "Cart closes," "last chance," "don't miss out," "only X spots left" (unless literally true). "Hey [first name]," — start with the hook, not the greeting.
What you get: A launch sequence that grows your list's trust even from people who don't buy.
What to do next: Schedule the sequence. Watch click-through rate, not just sales. CTR drop = email failed.
CATEGORY 4: RESEARCH & DECISIONS
Most solopreneurs don't have a research problem. They have a research avoidance problem. They guess at keywords, guess at competitor positioning, guess at what customers actually want — because real research used to take 6 hours per project. The Big Three just collapsed that to 6 minutes. These four prompts turn AI into your research department.
This is where Perplexity Computer pulls ahead of every other tool — because most of these prompts require live web data, not training data. (Here's how to actually run Perplexity Computer effectively — it's the agentic research layer that powers Category 4.)
From The Wolf Is at The Door: "The solopreneur who outresearches their market wins the market. AI didn't change that rule — it just made it survivable."
Prompt #14: The Striking-Distance SEO Rescue
Best in: Perplexity Computer (live SERP analysis + Search Console connector in one workflow)
You already have pages on your site ranking on position 5-15 for valuable keywords. They don't need new content — they need a targeted shove onto page 1. This prompt finds them, prioritizes them, and writes the exact optimization plan for each one. It's the highest-leverage SEO work most solopreneurs ignore.
Pro tip: This prompt requires live Search Console data, which is exactly what Perplexity Computer was built to handle — connect your Search Console and let it pull the data in-session.
THE PROMPT:
You are an SEO strategist trained in Aleyda Solis's content optimization methodology (rank improvement on existing pages is 10x ROI vs new content) and Eli Schwartz's product-led SEO framework (pages must rank AND convert, never just rank). I'll provide my Google Search Console export (Queries + Pages CSVs). Phase 1 — Find the rescue candidates: Identify every page with at least one keyword ranking position 5-15 AND at least 100 impressions in the last 90 days; For each, identify the highest-impression keyword sitting in that range (the rescue target); Output the top 10 rescue candidates ranked by combined impression volume + commercial intent. Phase 2 — Diagnose each candidate (do this for the top 3): Pull the current page URL; Pull the top 3 URLs currently ranking position 1-3 for that keyword; Identify what those 3 have that mine doesn't: depth, structure, freshness, specific subtopics, internal links, schema markup; Identify what mine has that they don't (where I'm uniquely strong). Phase 3 — Build the optimization plan for each (top 3 only): Headline + meta description rewrites; 3-5 sections to add (each tied to a subtopic the top 3 cover but I don't); 5 specific internal links I should add (with anchor text and source pages); Schema or structural changes; The single highest-impact change (do this first). Self-critique: which candidate has the highest projected ranking lift for the lowest effort? Tell me to start there. Banned: "Improve your SEO," "add more keywords," "make it more comprehensive." Every recommendation must be specific to the candidate page.
What you get: A ranked rescue list and surgical optimization plans for your top 3 winnable pages.
What to do next: Ship the highest-impact change today. Re-rank in 14 days. Move to candidate #2.
Prompt #15: The Competitor SERP Teardown
Best in: Perplexity Computer (live SERP + competitor site fetch in one session)
You don't need a generic "competitor analysis." You need to know exactly how 3 specific competitors are beating you on the keywords that matter — and what they're doing structurally that you can copy in 30 days. This prompt is the teardown.
THE PROMPT:
You are a competitive intelligence analyst trained in April Dunford's competitive positioning framework (your customer always has alternatives — name them) and Andy Crestodina's content competitive analysis methodology (you don't compete with your category, you compete with whoever's ranking). I'll name 3 competitors + 5 keywords I care about. Step 1: For each of the 5 keywords, pull the current top 10 ranking URLs. Tell me which competitors of mine appear, and which keywords they're winning on. Step 2: For each of the 3 named competitors, analyze their: Top 5 highest-traffic pages (best guess from public data); Their dominant content angle (educational, contrarian, story-driven, listicle, etc.); Their email lead magnet strategy (what they're capturing emails with); Their offer ladder (free → low ticket → main → high ticket); Their dominant social channel and what's working there. Step 3: Build the gap analysis: The 3 angles all 3 competitors are missing (where I can dominate); The 3 angles 1+ competitor is doing that I should counter; The 1 keyword cluster where I can realistically win position 1-3 in 90 days. Step 4: Build my 30-day counter-positioning plan: The 1 piece of pillar content I should ship in week 1; The 3 supporting pieces in weeks 2-4; The internal-link structure connecting them; The lead magnet positioning that differentiates from all 3 competitors. Self-critique: am I recommending I out-compete on their terms, or differentiate on terms they can't easily copy? If too "me-too," rewrite the plan. Banned: "Be better than them," "stand out from the competition," generic positioning advice. Every recommendation must be tied to a specific competitor move.
What you get: A 30-day plan to systematically take SERP territory from named competitors.
What to do next: Execute week 1. Track keyword movement. Iterate based on real ranking shifts.
Prompt #16: The Voice-of-Customer Goldmine
Best in: Claude (best at synthesizing 50+ review excerpts into structural patterns)
Your customers have already told you exactly how to sell to them. The words are sitting in your DMs, your email replies, your Amazon reviews, your YouTube comments, your competitor's review sections. The top 1% mine this language and use it verbatim in their copy. The bottom 50% write copy from their own head. This prompt does the mining.
THE PROMPT:
You are a voice-of-customer researcher trained in Joanna Wiebe's customer-language methodology (the best copy isn't written — it's stolen verbatim from your customers) and Indi Young's listening research framework (people's actual problems live in their own words, never in your category language). I'll paste 30-50 verbatim customer excerpts (DMs, emails, reviews, comments, support tickets, competitor reviews). Run this exact 5-phase mining: Phase 1 — Pain language extraction: Pull every distinct pain point. List each one in the customer's exact wording (no paraphrase). Cluster identical pains across multiple customers. The 3 most frequently mentioned pains are your gold. Phase 2 — Desire language extraction: Pull every distinct desired outcome, in customers' exact wording. Identify the top 3 most-frequently-mentioned outcomes. Phase 3 — Objection language extraction: Pull every objection, hesitation, or "but" statement. Identify the top 3 most common objections. Phase 4 — Identity language extraction: Pull every phrase where the customer describes themselves ("I'm the kind of person who..."). Pull every phrase where they describe who they're NOT. These are gold for "you're the type of person who..." positioning. Phase 5 — The Swipeable Lines: Pull the 10 most quotable, specific, emotionally-loaded sentences across the entire dataset. These go in your sales page, your headlines, your subject lines — verbatim. Output everything as a single structured doc I can paste into my brand voice file. Self-critique: am I paraphrasing anywhere? If yes, replace with the verbatim quote. Banned: Summarizing in my words. Every output line must be a direct quote or a clear quote-attribution. If a customer didn't say it, it doesn't go in the doc.
What you get: A swipe file of your customer's exact language — the raw material for every piece of copy you'll write for the next year.
What to do next: Replace 5 phrases on your sales page with verbatim customer quotes from this doc. Watch conversion move.
Prompt #17: The Positioning Audit That Saves a Brand
Best in: Claude (handles competitive landscape reasoning at depth)
Most "positioning" exercises produce a slide deck nobody reads. The top 1% positioning audit identifies the single most expensive sentence on your homepage — the one losing you money right now — and fixes it. This prompt is that audit.
THE PROMPT:
You are a positioning strategist trained in April Dunford's "Obviously Awesome" methodology (positioning = where your product is uniquely best for a specific customer with a specific alternative) and Marty Neumeier's "Zag" framework (the strongest brands aren't the best — they're the only). I'll paste my homepage copy + my top 3 competitor homepages + my ICP description + my offer details. Run this exact 6-step audit: Step 1 — The "what we are" test: What category does my homepage claim I'm in? Is that the best category for me to be in, or just the obvious one? Compare to my 3 competitors' category claims. Step 2 — The "for whom" test: Who does my homepage say I'm for? Is the specificity tight enough that someone NOT in that group will self-deselect? If too broad, propose a tighter ICP. Step 3 — The "vs what alternative" test: What does my homepage position me against? "Doing nothing"? A specific tool? A category of solutions? If the homepage doesn't name the alternative, that's a critical gap. Step 4 — The "unique value" test: What does my homepage claim I uniquely deliver? Could any of my 3 competitors copy that claim word-for-word? If yes, the claim isn't unique. Step 5 — The "proof" test: What proof does the homepage give for the unique value claim? Specific numbers? Named customers? Concrete transformations? Or vague social proof? Step 6 — The single most expensive sentence: Identify the one sentence on my homepage that's losing me the most money. Tell me why. Rewrite it. Output: Current positioning score (1-10 per dimension); The 3 highest-leverage changes (ranked by projected impact); The exact rewrite of the single most-expensive sentence; A new positioning statement (one sentence) using the Dunford format: "For [ICP], [product] is the [category] that [unique value], unlike [alternative]". Self-critique: is my new positioning genuinely defensible or aspirational? If aspirational, ground it in what's actually true today. Banned: "We help businesses grow," "we're passionate about," "we empower," any homepage claim that survives the "could my biggest competitor say this too" test.
What you get: A surgical positioning fix and a rewritten homepage cornerstone.
What to do next: Ship the one-sentence rewrite today. Measure inbound lead quality over 30 days.
CATEGORY 5: BUSINESS OPERATIONS
This is the unsexy work that drains 10 hours from your week — contracts you don't understand, brand partnerships you under-negotiate, pricing decisions you make from your gut. It's also where AI quietly creates more leverage than anywhere else, because these are document-heavy, judgment-light tasks built for a reasoning model.
These three prompts turn AI into your business operations layer. They're the closest thing to having a fractional COO, lawyer, and pricing strategist on call. (Discussed at depth in my breakdown of the best AI books for 2026 — these prompts pair well with reading.)
From The Wolf Is at The Door: "The irreplaceable solopreneur isn't the one who works the most hours. It's the one who delegated the document work to AI two years before everyone else."
Prompt #18: The Contract Red-Flag Forensic Scanner
Best in: Claude (long context for 50+ page contracts + strongest legal reasoning)
You're signing contracts that contain landmines you can't see. Auto-renewal clauses, exclusivity creep, IP assignment buried in section 14, indemnification that nukes your business if a client claims damages. The top 1% don't pay a lawyer $500/hour for first-pass review — they run this prompt first, then send only the flagged items to the lawyer. Most contracts come back with 5-10 red flags.
AI legal review is a starting point — not a substitute for an attorney on anything significant.
THE PROMPT:
You are a contract risk analyst trained in the deal-protection frameworks of Bobby Klinck (solopreneur-specific contract risks: IP, exclusivity, payment terms, termination) and the contract-review methodology of Verbit-trained legal ops (every clause is scored on risk to the signer). I'll paste/upload a contract. Run this exact 7-zone scan and output a single risk-ranked table: Zone 1 — Payment terms: When am I paid? (Net 30/60/90?) Is there a late-payment penalty in my favor? Are there clawback provisions if a client cancels? Are kill fees specified? Zone 2 — Scope & deliverables: Is the scope tightly defined or "and other duties as assigned"? Are revisions limited (e.g. "2 rounds") or unlimited? Are deadlines mutual or one-sided? Zone 3 — Intellectual property: Who owns work product? When does ownership transfer? Is there language assigning my pre-existing IP to the client (this is the #1 trap)? Can I use the work in my portfolio? Zone 4 — Exclusivity & non-compete: Are there non-compete clauses? Geographic scope, duration, and category scope of any exclusivity? Does it survive contract termination? Zone 5 — Termination: Can either party terminate? On what notice? What happens to unpaid work on termination? Are there "cure periods" before termination? Zone 6 — Liability & indemnification: Is liability capped (and at what level)? Am I indemnifying the client against third-party claims? (This can nuke your business.) Is there mutual indemnification? Zone 7 — Renewal & assignment: Does the contract auto-renew? Can the other party assign the contract to another company? Are there change-of-control provisions? For each flag, output: Zone | Clause # | Quoted language (verbatim) | Risk level (Critical / Warning / OK) | Plain-English explanation | Suggested redline language. Then output: The top 3 must-fix-before-signing items; The 5 nice-to-fix items; The 2 sentences I should send to a lawyer for human review. Banned: Telling me "consult a lawyer" without first doing the analysis. Vague risk descriptions. Skipping clauses because they "look standard."
What you get: A risk-ranked breakdown of every dangerous clause + suggested redlines.
What to do next: Send only the Critical items to your lawyer. Negotiate the Warning items yourself.
Prompt #19: The Brand Partnership Negotiation Architect
Best in: ChatGPT (best at multi-turn negotiation scenario reasoning)
Most creators take the first number a brand offers. The top 1% never do — because they've already calculated their floor, their ceiling, the brand's likely budget, and the 3 leverage points they can negotiate beyond the dollar figure. This prompt builds the entire negotiation strategy before you reply to the email.
THE PROMPT:
You are a creator-economy deal strategist trained in Chris Voss's "Never Split the Difference" tactical empathy framework (label emotions, mirror, calibrate questions to force the other side to solve your problem) and Justin Welsh's solopreneur leverage model (your audience is the asset — price every component, not just the deliverable). I'll paste: the brand's outreach email + their offered terms + my audience stats + my past partnership rates + my non-negotiables. Run this exact 6-step deal architecture: Step 1 — Decode the brand's position: What's their likely budget range based on company size, vertical, campaign type? What's their actual goal (awareness, attribution, content licensing, exclusivity)? What are their constraints (timeline pressure, internal approvals, fiscal-year urgency)? What signals are in the email (urgency words, ambiguity, who's CC'd)? Step 2 — Calculate my real worth on this deal: My CPM-equivalent if this were a paid ad placement to my audience size; My deliverable-stack value (video + post + story + email + repurposing rights); My exclusivity premium if they're asking for category exclusivity; My time cost + opportunity cost. Step 3 — Find the value beyond cash: Could I negotiate product, equity, affiliate rev share, ongoing retainer? Could I negotiate content reuse rights on my own channels? Could I negotiate case study rights for my own marketing? Could I negotiate a smaller pilot deal with a clear scale-up path? Step 4 — Set my number triangle: Floor: the absolute minimum I'd accept; Target: what I expect to settle at; Ask: the opening number that anchors high but isn't absurd. Each must be justified with logic the brand will accept. Step 5 — Draft 3 reply versions: Version A (high-leverage): I have other offers or strong demand. Anchor at the ask. Version B (collaborative): I want to make this work. Anchor at the ask but pair with "let's find the structure."; Version C (deal-saving): They have a tight budget. Trade scope for value-adds, hold the line on dollars per deliverable. Step 6 — Predict their pushback + script my response: The 3 most likely pushback responses; My scripted response to each (using Chris Voss tactical labels and calibrated questions). Self-critique: am I leaving money on the table anywhere? Am I being greedy in a way that kills the deal? Identify the weakest move and rewrite it. Banned: Generic negotiation advice. Every recommendation must be tied to specific data from the brand's email and my situation.
What you get: A complete negotiation playbook with scripted replies and pushback responses.
What to do next: Send Version A or B. Track which pushback materializes. Use the scripted response.
Prompt #20: The Pricing & Offer Stack Architect
Best in: Claude (best at multi-variable pricing reasoning + offer-stack design)
You're underpricing. Almost every solopreneur is. The reason isn't because you're bad at pricing — it's because you priced your offer the day you launched it, and then you never re-architected it as your authority grew. This prompt rebuilds your offer stack from scratch using current market data and what the top 1% of your competitors charge.
Pro tip: Run this in Perplexity Computer so it can pull live competitor pricing pages in the same workflow.
THE PROMPT:
You are a pricing strategist trained in Hermann Simon's value-based pricing methodology (price is set by perceived value, not cost or competitor benchmarking alone) and Madhavan Ramanujam's "Monetizing Innovation" framework (offer architecture and bundling drive more revenue than headline price). I'll paste: my current offer + current price + my ICP + my top 5 competitors' pricing pages + my last 30 days of sales data. Run this exact 7-step architecture: Step 1 — Market position diagnostic: Where am I priced relative to competitors? (Lowest / mid / highest); Is my price signaling commodity or premium? (Specific to my market — what's the threshold?); What does my current price say about who I'm for? Step 2 — Value architecture diagnostic: What outcome does my customer get? Quantify it in dollars, hours, or status if possible; What's the gap between my price and that outcome value? (10x rule: price should be 1/10 of the value delivered, minimum). Step 3 — Willingness-to-pay signal mining: From competitor pricing, what's the market's anchor range? From my sales data, what's my price-elasticity signal (do refunds spike at certain price points, does conversion drop sharply)? From my customer research, what would they have paid? Step 4 — Re-architect the offer stack: Tier 1 (entry): The "no-brainer" offer that's 1/10 of the main offer's price. What's in it? Who's it for? What's the bridge to Tier 2? Tier 2 (main): The core offer, repriced for premium positioning. What's the stack inside it? What's the implied value of each component? Tier 3 (high-touch): The 5x-priced version. What human/done-for-you/access component justifies it? Tier 4 (top-of-pyramid): The 10x version. Who exactly is this for? (If you don't have a Tier 4, you're capping your business at Tier 2 customers.) Step 5 — Price each tier with justification: Specific dollar number for each; The "anchor" you'll show on the sales page (the comparison that makes the price feel small); The 5 components in each tier and the implied dollar value of each. Step 6 — Migration plan: How do I introduce the new prices to my existing audience without burning trust? How do I grandfather current customers? How do I sequence the rollout (email, sales page, ads)? Step 7 — The single highest-leverage change: The one move I should make this week (and why it's higher leverage than the others). Self-critique: am I pricing for my current authority or my actual delivered value? If those don't match, identify the gap and propose how to close it within 90 days. Banned: Generic pricing advice. "Charge what you're worth." Every recommendation must be tied to a specific dollar figure and a specific competitor or value benchmark.
What you get: A re-architected 4-tier offer stack with justified prices, anchors, and a rollout plan.
What to do next: Ship Tier 1 this week. Test conversion. Roll up to Tier 2 in 30 days.
Wolf at the Door: Why This Window Closes Fast
For ten years, the moat that protected the $100K marketing team was specialized knowledge — strategists knew positioning, copywriters knew conversion, SEO leads knew search algorithms, analysts knew funnels. Each person held a piece of the puzzle.
In a single week of June 2026, OpenAI, Google, and Anthropic collapsed every one of those puzzle pieces into a $20/month subscription. The moat didn't get smaller. It got vaporized.
This is the wolf at the door. Not for the marketing team — for you. Because the same prompts you just read? Your competitors are reading them too. The advantage isn't that AI can do the work. The advantage is that you use it before the rest of your market figures out it's possible.
The irreplaceable solopreneur in 2026 isn't the one who works the hardest or has the biggest budget. It's the one who built the sharpest prompt stack, the cleanest workflow, and the deepest customer understanding — and ships faster than anyone in their niche.
That's what The Wolf Is at The Door is about. The book lays out the full strategic framework for staying irreplaceable while the giants consolidate everything around you. These 20 prompts are the tactical layer. The book is the strategic one.
Three next steps:
Grab the Free AI Success Kit — the starter pack for solopreneurs running their business on AI.
Read The Wolf Is at The Door — the full playbook for staying irreplaceable in the age of disruption.
Bookmark this post. Re-run one prompt a week. In 20 weeks, you will have systematically rebuilt every layer of your marketing — and you'll have done it for the cost of a few AI subscriptions, not the cost of a marketing team.
The giants are coming for the marketing team. The question is whether you're the solopreneur who beats them to it.
Comments