Waffle

What do they actually do

Waffle is a consumer mobile app that teaches languages through spoken practice with an AI tutor. Users pick a target language and a roleplay scenario, speak with the AI in natural conversations, and get corrective feedback and suggestions for improvement. The product emphasizes speaking from day one, a library of roleplays, personalized lessons, and detailed feedback Waffle homepage, App Store.

The app is live on iOS and Android, with downloads available via the official store listings. Sessions can be reviewed, and the app tailors daily practice to the learner’s level and interests App Store, Google Play. Waffle is listed as an active company in Y Combinator’s Spring 2025 batch YC profile.

Who are their target customer(s)

• Casual learners without conversation partners: They want quick, guided speaking sessions they can do on their phone, because passive study isn’t building real speaking confidence. Waffle’s spoken roleplays and daily prompts target this gap homepage, App Store.
• Students or test‑takers preparing for oral exams/interviews: They struggle with nerves, spontaneous replies, and getting precise feedback on what to improve. Waffle’s conversation roleplays plus post‑session feedback address this need homepage, App Store.
• Working professionals using a second language at work: They need practical, context‑specific practice for meetings, calls, and emails, and quick phrasing/pronunciation help without blocking workflow. The founders have discussed an on‑screen assistant for live, in‑context feedback founder demo.
• Busy learners seeking a daily routine: They lose momentum without short, tailored lessons and reminders. Waffle promotes daily, personalized lessons and stored/reviewable past sessions to maintain habit and progression homepage.
• Non‑native speakers focused on pronunciation/fluency: Solo study rarely provides actionable corrections on how to sound more natural. Waffle centers on spoken conversations with detailed feedback to improve pronunciation and fluency App Store.

How would they acquire their first 10, 50, and 100 customers

• First 10: Use founders’ networks and YC warm intros to onboard friends, ex‑colleagues, and early adopters with free pro access, plus 1:1 onboarding calls to observe sessions and fix immediate UX/audio issues YC profile, App Store.
• First 50: Seed communities with short demo clips and honest practice stories in language subreddits/Discord/Facebook, distribute time‑limited promo codes, and A/B test “speak now” vs. “pass oral exams” messaging based on repeat‑session behavior.
• First 100: Run tightly targeted TikTok/Instagram creatives showing real voice roleplays and a one‑tap trial, optimizing for trial→retention; in parallel, offer short pilots to university language departments/tutoring networks for concentrated, high‑engagement cohorts and testimonials.

What is the rough total addressable market

Top-down context:

Global online language learning is roughly USD 21–22B in 2024–2025, per market reports Grand View Research, Mordor Intelligence.

Bottom-up calculation:

Assuming ~60–63% of online revenue is via mobile/self‑learning apps yields a mobile TAM of about USD 13–14B. If speaking/conversation features represent ~10–30% of that, the speaking‑focused SAM is roughly USD 1.3–4.2B Grand View Research, Mordor Intelligence.

Assumptions:

• Mobile/self‑learning represents ~60–63% of online language learning revenue today.
• Speaking/conversation features account for 10–30% of mobile app revenue.
• Waffle’s near‑term focus is consumer self‑study on mobile rather than corporate contracts.

Who are some of their notable competitors

• Cambly: On‑demand human tutors for live video/audio conversation practice; strong for real‑human interaction and corrections, but higher cost per minute vs. always‑available AI.
• ELSA Speak: Pronunciation‑focused coach with phonetic scoring and drills; narrower than Waffle’s full conversational roleplays and lesson flows.
• Duolingo: Mass‑market gamified lessons with speaking prompts and AI chat; breadth and engagement are high, but speaking feedback is typically lighter compared with targeted post‑session corrections.
• HelloTalk: Peer language exchange for text/voice chats with native speakers; offers real conversations but relies on community partners rather than a structured AI‑led curriculum.
• Speechling: Recorded speaking practice with coach or automated feedback and structured plans; closer on feedback depth, but oriented around recordings versus live, interactive roleplays.