Product Roadmap — WiderWings Inc

Last Updated: 2026-02-11
Owner: Product Manager (currently Bob until PM agent is active)

---

Vision

Build a billion-dollar enterprise in 5 years with AI agents as employees.

Year 1 Goal: Mid 6-figure revenue from MedSchools.ai

---

Current Focus: MedSchools.ai

Now (Feb 2026)

• Core platform live on Vercel
• UCLA POC with personalized AI advisor
• Mobile-first dashboard redesign
• Interview module with Vapi integration
• User onboarding flow polish
• Content system activation (blog, guides)

Next (Mar 2026)

• Billing/subscription integration (Stripe)
• Expand school coverage (beyond UCLA POC)
• SEO content push (school pages, rankings)
• User feedback collection system

Later (Q2 2026)

• Secondary essay AI assistance
• Application tracking features
• Mobile app consideration
• B2B/counselor offering exploration

---

Product Pipeline

MedSchools.ai → CollegeDojo.com → Real Estate Vertical

Each product follows the same playbook:

1. Personalization engine (user data → tailored advice)
2. Content moat (SEO + community data)
3. Subscription revenue

WiderWings.com (Infrastructure)

Marketing hub powering all products:

• Customer database
• Personalization engine
• Remarketing (email/SMS/phone)
• Paid media platform

---

Strategic Moats

1. Personalization — Use actual user data (MCAT, GPA, school list) to give advice ChatGPT can't
2. Fresh community data — SDN/Reddit insights that LLMs don't have
3. Vertical integration — Own the full stack from data to distribution

---

Key Decisions Log

Date	Decision	Rationale	Owner
2026-01-29	Personalization > RAG	Generic RAG loses to ChatGPT. Moat is user-specific data.	Henry + Bob
2026-02-05	Vapi for interviews	More responsive than OpenAI Realtime, flexible voice options	Bob
2026-02-10	Mobile-first dashboard	60%+ users on phones, prioritize mobile UX	Bob
2026-02-11	Org restructuring	Bob delegates more, multi-agent coordination	Henry

---

Resources

• Specs: ~/company/specs/
• Research: ~/company/research/
• User Feedback: ~/company/user-feedback/