Replacing World-Class Advisors with an Always-On AI Board
I’ve been lucky to have incredible advisors in past companies. But advisors aren’t always available. And decisions can’t wait. Here’s how I now use AI as a 24/7 board of directors.
I have been fortunate to learn from some of the best minds in the startup / busines world, including (above from left to right) Mark Roberge, Steve Anderson, John Keister, and Meagen Eisenberg.
Here is the reality. Advisors are incredible, but they cannot always pick up the phone. Your toughest questions rarely arrive at a convenient time. They certainly do not wait for people to be free.
Now that I am building SuperSmall, I find myself wanting that same level of guidance. I want insight, pattern recognition, and someone or something that can challenge my thinking.
So I created something new.
An always-on AI Board of Directors: ChatGPT, Claude, and Gemini.
Three seats at the table.
Three perspectives.
Zero waiting.
And the results have been surprisingly helpful.
Why 3???
Each AI is strong in different ways, similar to real advisors. Using three gives me variety, fewer blindspots, and sharper decision making. The power of three feels very real. And like most important things in life, I always want at least three independent datapoints.
How I Run My AI Board
To keep each session consistent, I follow a simple system:
I use the same prompt for all three AIs.
I attach the same data. For this post, it was the CSV of our AI House event signups.
I sometimes ask the models to think like a specific advisor, for example “Answer like Mark Roberge” for sales questions.
I keep the original formatting so you can see the personality of each model.
Today’s “Board Topic” and Prompt
To begin this series, I asked all three AIs the following:
Prompt: “You are like Mark Roberge, with an incredible ability to glean insights especially from customer information and feedback. Based on the following signups for a recent event SuperSmall hosted at AI House, what would be your top 5 insights in stacked ranked order? And what are the next steps based on those insights? Please limit your response to 300 words.”
*I also attached the recent sign up data in a .CSV file into all three LLM’s along with the prompt.
Below are the three outputs: A, B, and C.
Read each one. Notice how the thinking changes even though the prompt and data were identical. Then place your vote for the one you found most helpful.
Your vote will help the entire SuperSmall community understand how to use AI as a practical, everyday advisory tool. I will also share which option I selected and why in a follow-up post.
Note: I did keep the formatting as much as possible although that is a tell on which AI model it is. However, I felt that formatting is impactful to how people digest insights, so I kept it. The poll is underneath all 3 options.
You can always add comments to this post provide more context for your choice.
Let the Hunger AI Games begin!
OPTION A
Looking at 200+ attendees, here are the top 5 insights:
1. Non-Technical Founders Need Production Pathways (~40% of attendees)
Founders and PMs explicitly want to “build MVPs without a technical co-founder” but struggle getting from prototype to production-ready apps.
Next Steps: Create “Prototype to Production” workshop series; develop production readiness checklists; partner with deployment platforms (Vercel, Supabase).
2. Enterprise Employees = Premium Side-Hustle Market
Heavy Microsoft, Amazon, Google, Meta representation. They’re not just learning—they’re launching side projects and have budgets.
Next Steps: Launch $49-99/month premium membership; create “corporate moonlighter” content addressing IP/compliance; offer “ship in 30 days” cohorts.
3. Community Demand is Overwhelming
75%+ want “ongoing community support”—this is a community business, not just events.
Next Steps: Implement weekly office hours; create accountability pods (4-5 people); build showcase platform for shipped products; launch paid membership immediately.
4. Vertical Clustering = Specialized Opportunities
Strong representation in healthcare, real estate, education, enterprise ops. Domain experts with customer access seeking technical leverage.
Next Steps: Create vertical-specific workshops and prompt libraries; facilitate industry cohorts; develop case studies by vertical.
5. Skills Gap: Databases, Deployment, Quality Control
Repeated mentions of uncertainty around “backend,” “databases,” “production quality,” and “scaling.”
Next Steps: Build templates for common patterns (auth, payments, databases); offer code review services; create definitive deployment guides.
Critical Action:
Your waitlist/declined ratio shows demand exceeds capacity. Create paid fast-track option and nurture sequence now. This is a premium audience with money and motivation—monetize accordingly.
=====
OPTION B
Here’s the tight 300-word version with clear insights and next steps.
Top 5 Insights (Stack Ranked)
1. Massive appetite for community
~80% responded “Very interested” in ongoing community support. This is unusually high and validates your thesis that people want structured accountability around AI, not just content.
Next Step: Launch a Founding Member community pilot. Keep it small, curated, and focused on monthly workshops + peer groups.
2. Your ICP is exactly who attended
Founders (65) and tech employees (50) dominated signups. These are the people most likely to build, test, and launch with AI—your perfect audience.
Next Step: Create two tracks:
Founder Track: validate → build → test
Tech Employee Track: side project → prototype → revenue
3. Attendees want to build and validate, not learn theory
Open responses overwhelmingly mention: “test ideas,” “build prototypes,” “launch apps,” “create passive income,” and “automate workflows.” Very little interest in abstract AI education.
Next Step: Position your workshops around outcomes: vibe-coding, demand tests, MVP building, and early revenue.
4. Motivation is split between fear of being left behind + desire for opportunity
People explicitly referenced staying relevant, leveling up, and turning ideas into something real. This confirms a mix of FOMO and ambition.
Next Step: Use messaging that blends urgency with empowerment:
“Don’t get left behind. Build something customers want.”5. Strong signals this event is a conversion funnel
Attendees wrote thoughtful, multi-sentence responses—an indicator of high intent. Many asked for support after the event.
Next Step: Run a 5-email post-event drip:
Thank you + slides
Top insights from attendee responses
Founder case study
Invitation to the Founder Series (early-bird price)
Invitation to the Community (founding pricing)
If you want, I can now break these insights into a deck, a landing page brief, or a drip sequence.
=====
OPTION C
As a growth leader, I see clear signals in this data for immediate monetization and product expansion. The customer feedback confirms exceptional Product-Community Fit and defines a highly valuable, action-oriented segment.
Top 5 Stacked-Ranked Insights
Community is the Next Product: 73% of respondents are “Very interested” in ongoing community support, representing your most direct path to recurring revenue and customer retention.
The Audience is the Builder Segment: Signups are dominated by Founders and Startup Operators (the highest-intent, highest-value segment), whose primary goal is to launch Startups/Side Projects.
Specific Methodology is the Core Differentiator: High keyword frequency for “Vibe” and “Tools” indicates attendees aren’t seeking general knowledge, but a specific, rapid implementation framework (Vibe Coding).
Demand Far Exceeds Supply: 85 individuals (over 40% of total signups) were waitlisted or declined. This is a large pool of highly-qualified, uncaptured revenue.
Commercial Outcomes Drive Enrollment: The customer intent is clearly focused on moving quickly from idea to shipped code and business viability, valuing practical application over academic theory.
Next Steps
Launch a Beta Paid Community/Membership: Immediately convert the 73% who expressed strong interest. Offer the waitlisted/declined leads special “Founding Member” access to monetize uncaptured demand.
Brand the IP: Formalize the methodology (“The Vibe Coding System”) to create a proprietary, defensible framework and unique selling point for all future courses and content.
Re-align Value Proposition: Update all marketing copy to emphasize “How to Ship Your Profitable AI Startup/Side Project,” ensuring the commercial outcome is the primary benefit communicated to your Founder audience.
OK - there you have it three choices: A, B, or C. Please vote below, and I’ll do a followup post on what the community decided and which one I liked best and why :)
Thank you for reading through the three options. Your feedback is incredibly valuable for the entire SuperSmall community. This experiment is meant to help all of us sharpen our thinking, reduce blindspots, and make better decisions with the support of an always-on AI board.
Please take a moment to vote for the option you felt delivered the clearest and most actionable insights. If you have additional thoughts, leave a comment so others can learn from your perspective.
I will review the poll results, share which option I chose, and explain why in a follow-up post. My hope is that this becomes a regular series where we explore real founder problems together using practical, everyday AI tools.
Let’s continue building smarter and faster as a community.
Check out our upcoming in-person events.
