business

Verdict

Submitted 5/26/2026, 4:47:50 PM · Completed 5/26/2026, 4:53:05 PM

5.5
pivot
The idea

I built an AI flashcard app that generates study decks from any topic in seconds. 4 months in, 200 users, need honest feedback

Show original source text →
Hey! Been working on this for about 4 months as a solo dev and wanted to share where I'm at. The app: Genius Flashcards. You type a topic like "French irregular verbs" or "Contract law basics" and AI generates a full flashcard deck in seconds. You can also scan photos of your notes or PDFs and it converts them to flashcards. Then spaced repetition handles the review scheduling. Tech stack: SwiftUI (iOS), Next.js PWA (web/Android via Capacitor), Firebase, Gemini API for generation, Stripe for web payments, StoreKit for iOS. What went well: * AI generation is what people use most. Nobody likes making flashcards — they like using them * Building web + Android from one PWA codebase saved months * Spaced repetition + streaks create real retention. People come back because they feel guilty breaking their streak What I got wrong: * Over-built before validating. Built friends, study groups, leaderboards before confirming anyone wanted the core product * Should have launched with just "type topic → get cards" and nothing else * Spent way too long on features, not enough on distribution Current numbers (being honest): * \~200 users * 2 paid subscribers (both churned) * Free tier: 5 AI generations, 10 sets * Pro: €6.99/week or €49.99/year What I want feedback on: * First impression when you land on the app/site? * Does the AI generation feel useful or gimmicky? * Would you pay for this? What would make you pay? * What's the one thing that would make you tell a friend about it? Try it here: * iOS: [https://apps.apple.com/app/id6758582680](https://apps.apple.com/app/id6758582680) * Web app: [https://geniusflashcards.app](https://geniusflashcards.app) (no account needed to browse) * Android: still in review on Play Store Appreciate any feedback, positive or brutal. Thanks
TRIZ inventive level: 3/5· Principles: parameter changes, self-service
Synthesis verdict
**Pivot**. The idea of Genius Flashcards has potential, but the execution and validation process could be improved. The solo dev has already built a functional app with AI-generated flashcards, spaced repetition, and multiple platforms (iOS, web, Android). However, they over-built before validating the core product, which led to a slow start with only ~200 users and 2 paid subscribers. The tech stack used is modern and suitable for the task, but the monetization path is broken. Pricing (€6.99/week or €49.99/year) is aggressive for a niche tool with low perceived urgency. The free tier (5 AI generations, 10 sets) is too restrictive to drive conversion, and the lack of a middle-ground option (e.g., €2.99/month) alienates price-sensitive users. Distribution is the bigger failure: 200 users after 4 months with no clear acquisition channel (e.g., student forums, language-learning communities) means the product isn’t reaching its audience.

Strengths

  • The AI generation feature is the standout value proposition, as evidenced by user behavior (most usage focuses on this).
  • The tech stack used is modern and suitable for the task.
  • The core idea of AI-generated flashcards with spaced repetition addresses a clear unmet need: students and professionals consistently dislike manually creating flashcards but require them for effective learning.

Weaknesses

  • The monetization path is broken, with aggressive pricing and a lack of proven monetization strategy.
  • The product was over-built before validation, leading to a slow start and weak user traction.
  • The reliance on a single AI API (Gemini) introduces platform risk, should costs increase or the API's terms change unfavorably.

Best angle

Focus on simplifying the initial product to the core feature, expanding the free tier to reduce friction, and testing lower price points to improve distribution and monetization.

Panel verdicts

Competition

nvidia/nemotron-3-nano-omni-30b-a3b-reasoning

7.0

The main value proposition is the combination of AI-powered flashcard creation with a user-friendly interface, addressing a clear need in the language learning and study space.

The idea has a clear purpose (business venture) and a specific product (Genius Flashcards) with a defined target audience (students, learners). Existing competitors include Anki, Quizlet, and Quizlet itself, which offer flashcard creation and spaced repetition. The differentiation is real: the AI generation of flashcards from topics, photos, or documents is a unique feature not commonly found in other tools. The product addresses a clear need (efficient flashcard creation) and has a defined target audience. However, the success of this idea depends on execution, market fit, and competition, which are not fully addressed in the provided information. The concept has potential but requires validation through market research and user testing.

Viability

meta/llama-4-maverick-17b-128e-instruct

6.0

Simplifying the initial product to the core feature and focusing on distribution could have led to faster validation and potentially more users.

The idea of Genius Flashcards is technically feasible and has shown some promising results, but the execution and validation process could be improved. The solo dev has already built a functional app with AI-generated flashcards, spaced repetition, and multiple platforms (iOS, web, Android). However, they over-built before validating the core product, which led to a slow start with only ~200 users and 2 paid subscribers. The tech stack used is modern and suitable for the task. A 2-person team could potentially rebuild or improve v1 in 4-12 weeks, focusing on the core 'type topic → get cards' feature and distribution. The main challenges would be refining the AI generation, implementing a robust payment system, and improving user retention. The current free tier and Pro pricing model may need adjustments to encourage more paid subscriptions.

Risk

nvidia/llama-3.3-nemotron-super-49b-v1

2.0

Aggressive pricing and lack of proven monetization strategy amidst weak user traction and platform dependency threaten the venture's survival within 6-12 months.

The venture's viability is severely threatened by its inability to monetize effectively, coupled with a flawed initial development approach and insufficient user traction. Despite a potentially useful AI-generated flashcard feature, the product's over-building before validation led to a bloated initial offering that distracted from the core value proposition. The current pricing model (€6.99/week or €49.99/year) is overly aggressive for a utility that, while convenient, may not be perceived as essential by enough users to drive significant revenue. With only ~200 users and no retained paid subscribers, the app lacks evidence of market demand at the proposed price point. Furthermore, the reliance on a single AI API (Gemini) introduces platform risk, should costs increase or the API's terms change unfavorably. Regulatory risks, though less immediate, could arise from data protection concerns, especially with the scanning of personal notes and PDFs. Churn is already evident with the two paid subscribers churning, indicating either the product's failure to deliver long-term value or pricing misalignment. The target market may also consist of students or individuals with limited budgets, making the high weekly/monthly fee a barrier.

Monetization

mistralai/mistral-medium-3.5-128b

4.0

Fix pricing tiers and distribution before adding features—no one pays for a tool they can’t discover or justify the cost of.

The core value proposition—AI-generated flashcards with spaced repetition—is strong, but the monetization path is broken. Pricing (€6.99/week or €49.99/year) is aggressive for a niche tool with low perceived urgency. The free tier (5 AI generations, 10 sets) is too restrictive to drive conversion, and the lack of a middle-ground option (e.g., €2.99/month) alienates price-sensitive users. Churn of both paid subscribers signals misaligned value-to-price. Distribution is the bigger failure: 200 users after 4 months with no clear acquisition channel (e.g., student forums, language-learning communities) means the product isn’t reaching its audience. The AI generation is the hook, but without a viral loop (e.g., shareable decks) or a freemium model that scales with usage, retention and monetization will stall. Unit economics are unsustainable: Firebase + Gemini API costs likely outpace revenue at this scale.

Market

mistralai/mistral-small-4-119b-2603(fallback #2)

7.0

AI-generated flashcards solve a real pain point, but the app must prove retention and value before scaling, with a sharper focus on distribution and pricing.

The core idea of AI-generated flashcards with spaced repetition addresses a clear unmet need: students and professionals consistently dislike manually creating flashcards but require them for effective learning. The AI generation feature is the standout value proposition, as evidenced by user behavior (most usage focuses on this). The tech stack is well-chosen for cross-platform efficiency, and the monetization model (freemium with subscription) aligns with common SaaS patterns. However, the current traction is weak: only 200 users and 2 paid subscribers (both churned) suggests either poor distribution, weak value demonstration, or misaligned pricing. The overbuilt features (study groups, leaderboards) likely diluted focus from the core product, and the free tier may be too restrictive (5 generations is low for meaningful testing). The willingness to pay is unclear—€6.99/week is steep for a niche tool, and annual pricing at €49.99 may deter casual users. The audience size is substantial: global students (1.5B+), professionals (e.g., lawyers, med students), and lifelong learners (e.g., language learners). However, the app must prove retention and conversion before scaling. Key improvements: (1) Simplify onboarding to just 'type topic → get cards', (2) Expand free tier to 20-50 generations to reduce friction, (3) Test lower price points (e.g., €2.99/week or €29/year), (4) Focus on organic distribution (e.g., Reddit, Discord study groups, TikTok/YouTube tutorials). The AI generation feels useful, not gimmicky, but the app must demonstrate tangible learning outcomes to justify payments.

Synthesized by meta/llama-3.3-70b-instruct · 10.2s