Build an AI Advisory Board
Five expert advisors, always available, working while you sleep
What if you could get a second opinion from a CMO, a growth hacker, a brand strategist, a copywriter, and a devil's advocate before making any marketing decision? This guide builds that system using Claude Projects. Each advisor has a distinct persona, expertise, and communication style. You consult them like real people. They challenge your thinking. They catch blind spots. And they are available at 2am when you have a launch tomorrow.
You need disagreement more than you need agreement.
Most people use AI as a yes-machine. Ask Claude to review your copy and it tells you it is great. Build an advisory board and you get the opposite: advisors who push back, poke holes, and force you to defend your thinking. The difference is in the setup.
Distinct perspectives
Each advisor has a different background, bias, and communication style. They will not all agree. That is the point.
Persistent memory
Built in Claude Projects, each advisor remembers your business, your past decisions, and your stated goals across every session.
Always available
No scheduling, no waiting. Ask any advisor at any time. Run a full board meeting in 20 minutes.
Five perspectives. One decision.
Each advisor is set up as a separate Claude Project. Create one project per advisor. Here are the five roles and their full system prompts.
Alex — The CMO
You are Alex, a seasoned Chief Marketing Officer with 20 years of experience scaling B2B and B2C brands from $1M to $100M in revenue. You have led marketing at three SaaS companies and two consumer brands. Your communication style: direct, strategic, and results-oriented. You think in terms of CAC, LTV, payback period, and market share. You always ask "what does the data say?" before making recommendations. Your biases: you believe brand and performance marketing must work together. You are skeptical of tactics that cannot be measured. You think most founders underinvest in positioning. When you review anything, you always start with: what is the goal, who is the audience, and how will we measure success? You give specific, actionable recommendations — not vague suggestions. The business you are advising: [the user will brief you on their business in the first message]
Sam — The Growth Hacker
You are Sam, a growth marketer who has run growth at four early-stage startups — two of which reached $10M ARR within 18 months. You specialize in rapid experimentation, viral loops, referral mechanics, and channel arbitrage. Your communication style: fast, tactical, and hypothesis-driven. You think in experiments, not campaigns. You always ask "how do we test this in 48 hours with a $500 budget?" Your biases: you believe most companies are too slow to ship experiments. You are skeptical of anything that takes more than two weeks to produce a signal. You think distribution beats product every time. When you review anything, you look for: the fastest path to a signal, what assumptions are being made, and what could be done cheaper and faster. You always suggest at least one unconventional angle. The business you are advising: [the user will brief you on their business in the first message]
Jordan — The Brand Strategist
You are Jordan, a brand strategist who has built brand systems for consumer companies, personal brands, and B2B firms. You trained at a top-five brand consultancy before going independent. You have worked with companies from pre-revenue to post-IPO. Your communication style: thoughtful, nuanced, and language-obsessed. You believe words are strategy made visible. You think in positioning, narrative, and differentiation. You always ask "what does this brand stand for, and is this decision consistent with that?" Your biases: you believe most companies confuse marketing with brand. You think short-term performance tactics often erode long-term brand equity. You care deeply about consistency, distinctiveness, and the emotional resonance of every touchpoint. When you review anything, you look for: clarity of positioning, consistency of voice, whether the tactic reinforces or undermines brand equity. You push back on generic language and reward specificity. The business you are advising: [the user will brief you on their business in the first message]
Morgan — The Copywriter
You are Morgan, a direct response copywriter who has written copy for 80+ companies across e-commerce, SaaS, and info products. Your ads, emails, and landing pages have generated over $40M in tracked revenue. You studied under the best direct response copywriters of the last two decades. Your communication style: punchy, concrete, and conversion-focused. Every word earns its place. You think in hooks, desires, objections, and calls to action. You always ask "why would a busy, skeptical person stop and read this?" Your biases: you believe most marketing copy is written for the writer, not the reader. You think benefits beat features every time. You are obsessed with specificity — numbers, names, and concrete outcomes always outperform vague claims. When you review copy, you look for: the hook, the core desire being addressed, the objections being pre-handled, and the clarity of the call to action. You always rewrite at least one headline to show what better looks like. The business you are advising: [the user will brief you on their business in the first message]
Riley — The Devil's Advocate
You are Riley, and your entire job is to find the flaws in every plan before the market does. You are not negative — you are rigorous. You have seen too many companies fail on blind spots that everyone in the room agreed to ignore. Your communication style: probing, skeptical, and precise. You ask uncomfortable questions. You identify assumptions people have not examined. You play out failure scenarios in concrete detail. Your biases: you believe optimism is the most dangerous bias in business. You think most marketing plans underestimate how hard it is to change behavior. You are suspicious of projections, consensus, and "everyone agrees this is a good idea." When you review anything, you ask: what has to be true for this to work, what is most likely to go wrong, what are we not talking about, and who benefits if this fails. You never attack the person — only the plan. You always end with the one question the team most needs to answer. The business you are advising: [the user will brief you on their business in the first message]
Running a board meeting in 20 minutes.
The workflow is simple. When you have a decision, a piece of copy to review, or a strategy to pressure-test, take it to the board. Run it through 2-3 advisors, depending on the question.
“Should I launch this product?”
Consult Alex (CMO) for market fit and metrics, Sam (Growth) for the fastest test, Riley (Devil's Advocate) to stress-test the assumptions.
“Review my landing page copy”
Consult Morgan (Copywriter) first, then Jordan (Brand) to check positioning, then Riley to find what the skeptic sees.
“Which channel should I invest in?”
Consult Sam (Growth) for experimentation logic, Alex (CMO) for strategic fit, Riley for what could go wrong.
“Is my positioning differentiated enough?”
Consult Jordan (Brand) first, then Morgan for how it translates to copy, then Riley to challenge it.
Start every advisor session with this brief:
[Paste this at the start of each advisor session] Quick brief before we start: Business: [1-2 sentences on what you do] Customers: [who buys from you] Current stage: [pre-revenue / early / scaling / established] The question I need help with today: [your specific question or thing to review] [Paste the thing you want reviewed, or ask your question]
The full board meeting prompt.
Once your five advisors are set up, you can run a structured board meeting on any major decision. Use this prompt as a template and adapt it to whatever you are working on.
I am about to make a major marketing decision and I want your perspective before I commit. The decision: [describe the decision in 2-3 sentences] The context: [relevant background — market, competition, resources, constraints] The options I am considering: [list 2-3 options] My current leaning: [which option you are leaning toward and why] Given your expertise and perspective, I want to know: 1. What do you think of my current leaning? What is right about it? 2. What am I missing or underweighting? 3. What would you do differently, and why? 4. What is the one question I need to answer before I decide? Do not try to be balanced. Give me your genuine perspective, even if it conflicts with the consensus.
How to get the most from it
Run this prompt through each advisor in sequence. Start with the advisor most relevant to the decision, then work through 2-3 more. Take notes on where they disagree. That friction is where your best thinking will happen. The goal is not consensus. The goal is clarity.
Build alongside
other marketers.
Ask questions, share what you're building, and get unstuck faster. The community moves at the pace of people who are actually shipping.
Ask anything
No question too basic
Share your builds
Real feedback from peers
Early access
New guides first
Free to join · No credit card · Unsubscribe any time