Generic 'generate text' AI tools produce mediocre output because users don't know how to prompt effectively, resulting in low-quality content.
A step-by-step AI writing tool that interviews users about their intent, audience, and structure before generating content — turning the writing process into a guided workflow.
Freemium with subscription for advanced workflows and higher usage
The pain is real — most people get mediocre AI output and blame themselves or the tool. But it's a 'vitamin not painkiller' risk: people tolerate bad AI output and manually edit rather than seeking a better workflow. The friction is genuine but not hair-on-fire urgent. Power users feel it most.
TAM is massive — tens of millions of content creators, marketers, and knowledge workers use AI writing tools. SAM is more focused: professionals who write frequently enough to pay for better tooling (~5-10M). SOM for a solo founder realistically 5K-50K paying users, which at $15-30/month is a strong indie business ($1M-$18M ARR ceiling).
This is the weakest link. Users already pay for Jasper/Copy.ai but switching costs are low and AI writing is becoming commoditized. Many users feel ChatGPT at $20/month 'should do everything.' Willingness to pay a premium specifically for the interview UX is unproven. Must demonstrate dramatic quality improvement to justify separate spend. B2B positioning (agencies, content teams) would improve WTP significantly.
Very buildable. Core is an LLM orchestration layer with a structured conversation UI — prompt chaining, conditional branching, and template management. No novel AI research needed. A solo dev with LLM API experience can ship an MVP in 3-4 weeks. Tech stack: React/Next.js frontend + any LLM API + simple database for workflows and user state.
No one owns the 'interview-first' writing UX. Jasper has templates (forms, not conversations), ChatGPT can be prompted to ask questions (but inconsistently), and no tool makes the pre-writing interview the core product experience. The gap is clear. However, any competitor could bolt on an 'interview mode' in weeks, so the moat must come from workflow quality, not just the concept.
Strong subscription fit. Writers produce content regularly (daily/weekly), workflows can be saved and reused, usage-based pricing on generations is natural. Adding team features, custom workflow builders, and workflow marketplace creates expansion revenue. Monthly active usage patterns support subscription well.
- +Clear differentiation in a crowded market — 'interview-first' is a memorable positioning that no one owns
- +Technically simple MVP with fast time-to-market; the value is in UX design and prompt engineering, not infrastructure
- +Addresses a real and growing frustration as AI writing tools commoditize on raw generation quality
- +Natural expansion into workflow marketplace, team templates, and vertical-specific packages (legal, medical, marketing)
- +Low CAC potential — the concept is demo-friendly and shareable ('watch it interview me then write a perfect blog post')
- !Commoditization risk: ChatGPT/Claude adding 'ask me clarifying questions first' as a default behavior could eliminate the core value prop overnight
- !Willingness-to-pay risk: users may see this as a feature of existing tools rather than a standalone product worth paying for separately
- !Retention risk: once users learn to prompt better (which this tool teaches them), they may graduate back to cheaper general-purpose tools
- !Quality ceiling: the output is still only as good as the underlying LLM — if the interview is great but generation is mediocre, users blame the product
- !Feature creep danger: pressure to support every content type will dilute the focused interview experience that makes it special
AI content platform with templates, workflows, and brand voice features for marketing teams. Offers pre-built templates that guide users through structured inputs before generation.
AI copywriting tool with workflow automation and structured content generation pipelines for GTM teams.
General-purpose AI assistants that can be prompted or configured to ask clarifying questions before writing. Custom GPTs can be built with interview-style system prompts.
AI-native writing editor focused on long-form content with inline AI assistance and a more thoughtful writing experience.
Niche content tools that offer some structure — Typefully for Twitter threads, Castmagic for repurposing audio, ContentBot for workflow-based content. Each adds mild structure to AI generation.
Week 1-2: Build 3 killer interview workflows — blog post, LinkedIn post, and cold email. Each workflow is a 5-8 question adaptive conversation that captures intent, audience, tone, and key points. Week 3-4: Add generation with 'revision interview' (AI asks 'what would you change?' instead of manual editing). Ship with usage-limited free tier (5 generations/month) and $15/month Pro. Focus obsessively on output quality for these 3 use cases rather than breadth.
Free (5 generations/month, 3 workflows) → Pro $15/month (unlimited generations, all workflows, revision interviews) → Team $30/user/month (shared workflows, brand voice, collaboration) → Workflow Marketplace (user-created workflows, 70/30 revenue split) → API/White-label for agencies and content platforms
4-6 weeks to MVP and first free users. 8-12 weeks to first paying customers if launched with strong demo content on Twitter/LinkedIn/Reddit. The interview format is inherently shareable — screen recordings of the interview-to-output flow could drive organic growth. First $1K MRR realistically achievable in 3-4 months with consistent content marketing.
- “found a super specific friction point in how people use AI for writing”
- “Instead of slapping a generic 'generate text' button on a page, he built this structured, interview-style AI tool”