7 REAL TIPS for using AI on Kickstarter

Key Concepts

• AI-Driven Crowdfunding: Using Large Language Models (LLMs) like ChatGPT, Claude, and Gemini to streamline the research, copywriting, and strategic planning for Kickstarter campaigns.
• Deep Research Mode: A feature in AI models that performs extensive web searches and data synthesis to analyze competitors.
• Messaging and Direction Summary: A foundational document that defines product positioning, unique selling propositions (USPs), and target audiences.
• Brand Voice & Style Guide: A set of instructions that dictates the tone, personality, and linguistic quirks of a brand to ensure consistency across marketing assets.
• Pre-launch Funnel: The strategy of collecting email addresses and refundable deposits to build momentum before a campaign goes live.
• Predictive Modeling: Using AI to generate interactive tools that forecast revenue and Return on Ad Spend (ROAS) based on pre-launch metrics.

---

1. Competitive Research

The speaker emphasizes using the "Deep Research" mode in AI models to analyze competitors.

• Methodology: Use a prompt specifying the product category (e.g., "10-in-1 camping utensil") and target platform (e.g., Reddit).
• Goal: Identify top competitors, what users like (durability, portability), and—most importantly—what users dislike.
• Actionable Insight: Lean into the "negatives" identified in competitor reviews to differentiate your product and solve existing market pain points.

2. Messaging and Direction Summary

This document acts as the "single source of truth" for all future marketing efforts.

• Process: Use a "Senior Brand Strategist" persona prompt to interview the creator.
• Components: Includes product pitch, features/benefits, materials, use cases, target audience, and pricing.
• Utility: Serves as a guide for both human team members and as a context-provider for LLMs when generating future marketing assets.

3. Brand Voice and Style Guide

To ensure consistent communication, create a style guide by analyzing existing successful brands.

• Methodology: Input the URL of a brand’s landing page into an LLM and request a detailed style guide (tone, personality, sentence structure).
• Application: Use this guide as a "system prompt" for AI to ensure all generated copy sounds like your specific brand.

4. Campaign Page Copywriting

Instead of designing first, write all copy in a document to ensure the narrative is sound.

• Framework: Combine the Messaging Summary, Style Guide, and a Campaign Page Template into a master prompt.
• Outcome: The AI generates a structured campaign page, including suggestions for imagery, which can then be refined by the human team.

5. Video Scripting and Visual Breakdown

AI can deconstruct successful competitor videos to provide a blueprint for your own.

• Process:
  1. Use "Inspect Element" in a browser to extract the direct .mp4 URL of a video.
  2. Upload the video to Gemini.
  3. Prompt the AI to provide a "visual script" that breaks down the voiceover alongside corresponding visual cues.
• Application: Use this breakdown as a starting point to write a script that adapts successful pacing to your unique product.

6. Ad Concept Brainstorming

Use Meta’s Ad Library to find long-running competitor ads (a sign of success) and iterate on them.

• Methodology: Upload a screenshot of a successful ad and an image of your product to ChatGPT.
• Prompting: Ask the AI to create a new ad concept in the same style but with copy tailored to your product’s messaging summary.
• Refinement: Use iterative prompting to remove excess text and emphasize specific features.

7. Predictive Financial Modeling

AI can build custom, interactive calculators to estimate campaign success.

• Process: Provide an LLM with a prompt defining your inputs (ad spend, lead count, conversion rates, price point) and desired outputs (projected revenue, ROAS).
• Result: The AI generates an interactive web widget where you can adjust variables in real-time to forecast outcomes.

---

Synthesis and Conclusion

The core philosophy presented is that AI should not be used to "automate" a campaign in 10 minutes, but rather to accelerate the process by 90-95%. By creating a robust foundation—specifically the Messaging Summary and Style Guide—creators can ensure that all subsequent AI-generated assets (ads, page copy, scripts) remain on-brand and strategically aligned. The ultimate goal is to use these tools to build a high-quality pre-launch funnel, which the speaker identifies as the most critical factor in a successful crowdfunding campaign.