Contextual GRE Vocabulary Coach

Part three

Working notes

The opportunity

Verdict

Conditionally viable as a narrow GRE product; weak as a broad vocabulary subscription. The need is real and measurable, but adaptive scheduling, generated questions, multimedia, and social content are all reproducible features. The business earns a go for paid validation, not a go for full build. Its best chance is to own one outcome: improving contextual recall for GRE candidates with a near-term score goal.

The original scope—SAT, GMAT, GRE, other tests, and language learning—mixes customers with different jobs, content standards, willingness to pay, and acquisition paths. GRE is the better starting wedge because ETS explicitly assesses relationships among words and concepts and uses Text Completion and Sentence Equivalence formats. SAT should wait: its Words in Context domain emphasizes meaning and rhetorical purpose inside passages, so a generic difficult-word deck would be poorly aligned. GMAT and language learning should not be assumed adjacent until separate discovery supports them.

Evidence and market bounds

• Verified: ETS reports 256,215 GRE General Test takers with valid, reportable scores in testing year 2023–24, down from 467,277 in 2019–20. This establishes a meaningful but contracting annual population and argues against venture-scale claims based on the GRE alone.
• Verified: GRE Verbal contains 27 scored questions across two sections; ETS describes Reading Comprehension, Text Completion, and Sentence Equivalence, including understanding meanings and relationships among words and concepts.
• Verified: Paid demand exists, but so does a strong price ceiling. Vocabulary.com lists individual plans from $59.99/year to $12.99/month; Quizlet lists annual plans at $35.99 and $44.99; Magoosh provides a free GRE vocabulary builder; ETS promotes 1,600+ official GRE vocabulary practice questions through Brainscape.
• Informed estimate: If 30% of annual GRE takers are reachable English-language digital-prep buyers, the initial SAM is about 76,865 candidates (256,215 × 30%). At a proposed $39 exam pass, that is roughly $3.0 million in annual gross revenue, not a market forecast. A first SOM target of 1% of that SAM is about 769 buyers / $30,000 annual gross revenue. The 30% and 1% inputs are assumptions that must be tested.
• Opinion: GRE vocabulary can support a focused, capital-efficient product, but likely needs later expansion into broader verbal reasoning, tutor licensing, or a second exam to become a large standalone company.

Competitive map

The proposed product cannot credibly claim an adaptive-learning advantage: Vocabulary.com, Quizlet, Anki, and Brainscape already adapt or schedule practice. Nor is unlimited generated content inherently valuable; unreliable questions can destroy trust. The differentiation hypothesis is validated transfer: a learner sees that mastery on one wording persists days later and generalizes to a novel, GRE-like context.

Assumptions requiring validation

1. Pain: candidates attribute a material part of their verbal-score gap to contextual vocabulary rather than reading strategy or reasoning.
2. Payment: at least 20% of qualified prospects will pay $39 despite credible free alternatives.
3. Learning: constrained, reviewed variants improve delayed contextual recall by at least 15 percentage points versus the learner’s current method.
4. Retention: an imminent test date plus visible mastery is enough to sustain four sessions per week through week four.
5. Distribution: short-form viewers convert into diagnostic completions, not merely views; creator/tutor acquisition can remain below roughly $12 per paid pass.
6. Quality and rights: original questions can be consistently unambiguous and test-relevant without copying protected test material or implying ETS endorsement.
7. Expansion: a word-level performance graph can later support adjacent verbal practice; this is an option, not current evidence.

Recommended changes

Validation plan

Run the following in sequence and stop when a gate fails. Total expected cash cost is $1,500–$4,000, excluding founder time; all figures are planning estimates.

1. Problem and workflow interviews — one week, $0–$300

Recruit 20 people taking the GRE within 12 weeks from Reddit/Discord communities, tutors, and campus groups. Ask them to screen-share their word lists, recent mistakes, study cadence, and purchases. Do not pitch until the end.

Pass: at least 12/20 actively study vocabulary; at least 8/20 show repeated difficulty transferring definitions into sentence context; at least 6 have paid for verbal prep or state a concrete budget. Fail: fewer than 8 demonstrate the transfer problem; pivot toward the more frequent verbal job rather than polishing vocabulary features.

2. Message and pre-sale test — two weeks, $300–$1,000

Publish three landing-page messages—“raise contextual recall,” “know exactly which words to review,” and “ten-minute daily GRE workout”—fed by 15 short-form posts and small creator placements. Offer a refundable $39 eight-week founding pass after a five-minute diagnostic.

Pass: 100 qualified diagnostic completions, at least 20 paid or deposit-backed commitments, and no channel above $12 cash acquisition cost per buyer. Track view → landing visit → diagnostic start → completion → purchase separately. Fail: fewer than 10 purchases after 100 qualified completions; interview non-buyers and change promise/segment before building.

3. Concierge learning pilot — four weeks, $700–$1,700

Give 25–50 learners a manually operated daily queue covering 100–150 word families. Use original, expert-reviewed items. Randomize equivalent weak words within each learner between the new contextual protocol and their current flashcard method; retest both with novel contexts after seven days.

Pass: at least 15 percentage points higher delayed accuracy for the contextual protocol, at least 40% of learners active four or more days in week four, fewer than 2% of questions reported ambiguous, and at least 70% of buyers unwilling to request the offered refund. Fail: no meaningful recall lift or poor retention; test a broader verbal-reasoning coach rather than scaling content.

4. Product beta and channel proof — six weeks, $500–$1,000

Automate only the diagnostic, review queue, explanations, payments, and analytics. Keep editorial review manual. Run one tutor cohort and one organic creator funnel.

Pass: 50 paid users, 40% week-four retained usage, 60% contribution margin before founder labor, and one channel capable of producing 25 paid users monthly at no more than $12 CAC. Decision: proceed to a 300-word-family product if all learning, retention, and channel gates pass; otherwise narrow or stop. Do not add SAT, GMAT, or language courses during this test.

Risks and controls

• Question validity: use a style guide, dual review for assessment items, distractor rationales, versioning, and an immediate report mechanism. Never use live or copied ETS questions.
• Misleading score claims: report word-level recall and practice performance until a sufficiently powered study links usage to score changes; avoid guaranteed score improvement.
• Model drift and cost: cache approved content and use generation for controlled variants, not real-time unconstrained assessment.
• Child privacy: the initial adult GRE segment reduces but does not remove privacy obligations. SAT expansion would add minors and require a separate privacy and safeguarding review.
• Platform dependence: capture email and diagnostic data with consent; do not make reach on TikTok, Instagram, or YouTube the sole durable asset.
• Seasonality and churn: the exam-pass model makes expected churn explicit; build referrals and tutor cohorts rather than hiding episodic demand behind annual subscriptions.

Sources

Accessed 2026-08-11.

1. ETS, A Snapshot of the Individuals Who Took the GRE General Test, July 2019–June 2024 — annual test-taker counts and trend.
2. ETS, Overview of the Verbal Reasoning Measure — skills and question formats.
3. ETS, Test Content and Structure — current section lengths and assessed abilities.
4. ETS, GRE vocabulary flashcards — official Brainscape offer and stated feature set.
5. College Board, 2025 SAT Suite Annual Report — more than two million 2025 SAT takers.
6. College Board, Digital SAT Suite Technical Manual — Words in Context design and vocabulary model.
7. Vocabulary.com, Learner subscription plans — current consumer positioning and listed prices.
8. Vocabulary.com, School subscription — adaptive engine, content scale, and test-prep positioning.
9. Quizlet, Upgrade plans — current paid tiers and feature limits.
10. Magoosh, GRE Vocabulary Builder — free leveled GRE practice.
11. Anki, Background — spaced-repetition approach and free clients.

Open questions

• Which verbal score range and target gain best identify buyers with an urgent vocabulary-specific gap?
• Can candidates accurately supply prior practice results, or must the diagnostic establish the baseline entirely in-product?
• Which creator or tutor has enough audience trust to test pre-sales without paid brand-building?
• What editorial rubric reliably distinguishes useful novel context from ambiguous generated noise?
• Does an eight-week pass outperform monthly pricing on conversion, completion, refunds, and perceived fairness?