← Back

EdTech Chrome Extension

01

Executive Summary

Problem → Solution

Problem: [Intermediate learners look a word up once and never see it again, so retention collapses inside a week.]

Solution: [A browser extension that converts any page a learner already reads into a spaced-repetition surface.]

My role: [Sole PM + builder — discovery, spec, prototype.]

Key Impact

[+N%]

[week-1 return rate vs. flashcard baseline]

Extension interface mockup
02

Market Opportunity & Business Case

Market Sizing

TAM  [N]M online language learners x $[X]/yr = $[X]B
SAM  [N]M desktop-first, English-reading learners = $[X]B
SOM  [X]% year-1 capture                          = $[X]M

Unit economics: [CAC $X via extension-store organic; LTV $Y at Z% annual retention → LTV:CAC of ~[3.1]x.]

Persona & Pain

[The Self-Directed Reader]: [reads in-target daily, refuses to maintain a separate flashcard app.]

03

Regulatory & Strategic Constraints

Key Risks / Policy

  • — GDPR: page content is user data — process locally, never log URLs.
  • — Store policy: broad host permissions trigger manual review.
  • — [IP: publisher text cannot be persisted server-side.]

Strategic Trade-off

A — Local-only

Chosen

[All highlights and review queues stay in browser storage.]

Upside

  • + No GDPR processor duties
  • + Instant store approval

Cost

  • − No cross-device sync
  • − Data lost on profile reset

Why I chose it: [Local-only storage kept the extension inside the store's low-risk permission tier and removed the GDPR processing burden entirely — worth losing cross-device sync in V1.]

04

Product Specification & Design

User Flow

  1. 01[Install → pick target language]
  2. 02[Highlight a word on any page]
  3. 03[First review prompt within 24h]
  4. 04[Streak surfaced on new-tab]
[Embed Figma prototype here]

What I Built

[Content script, spaced-repetition queue, review popup — TypeScript + Chrome APIs.]

Scope cut: [audio pronunciation deferred to V2.]

05

Execution & Metrics Framework

Metric Tree

North Star: weekly words retained / user
├─ Highlights per active week
├─ Review completion rate
└─ Week-4 active rate
Counter-metrics: prompts dismissed, uninstall rate

GTM

[Seed in language-learning subreddits and Discords, then store-SEO on long-tail queries. No paid spend in V1.]

06

Retrospective & Next Steps

What I'd Change in V2

  • Instrument earlier: [I shipped before defining the counter-metrics.]
  • Narrow the ICP: [beginners diluted the retention signal.]
  • Test the paywall: [willingness-to-pay is still unvalidated.]
  • Revisit sync: [with a DPA in place, cloud sync is now worth it.]