The Common Admission Test (CAT) is the gateway to India's IIMs, the most prestigious management institutions in the country. Roughly 300,000 graduates sit for it every year, chasing a few thousand seats. The pass-to-seat ratio makes it one of the most competitive exams in the world. Unlike the SAT or GRE, CAT prep is not a supplementary industry but a primary occupation for hundreds of thousands of aspirants, often for 12–18 months at a stretch.
The exam has three sections: Quantitative Ability, Data Interpretation & Logical Reasoning, and Verbal Ability & Reading Comprehension (VARC).
CAT VARC is where the India's sharpest graduates consistently underperform because nobody built the right practice tool for them.
VARC is widely considered the most anxiety-inducing. Unlike Quant, it does not reward formulaic preparation. It rewards daily reading, pattern recognition, and deliberate practice — and most aspirants do not get enough of either.
Why I built this
The starting question was simple: will people pay for a focused, well-made VARC practice app? Not will they download it, will they actually pay. That was the hypothesis worth testing. I had seen too many EdTech products drown in content production and never validate willingness to pay. I wanted to know the answer before I wrote a single line of content.
I chose VARC specifically because it needed daily practice and daily reading, which made it defensible. A quant app can be gamed in a weekend cramming session. VARC cannot. That creates a recurring use case, which creates retention, which creates a subscription-worthy product.
I also made a deliberate niche bet: build for one section only, and build it well. The existing CAT prep landscape is dominated by bloated platforms with hundreds of hours of video content, weak mobile experiences, and no real feedback loops. Going section-specific let me be genuinely good at one hard thing instead of mediocre at many.
Revenue — people did pay
The most important validation for any consumer EdTech product is not downloads. It is whether someone opens their wallet. VARC for CAT has crossed that threshold.
Revenue to date
| Category | Amount |
|---|---|
| First two subscribers (early cohort) | $4.00 |
| Current active subscriber | $2.00 |
| Total collected | $6.00 |
The dollar amounts are small. That is the point. This was never about revenue at this stage, it was about willingness to pay. Three separate people, without any paid acquisition, without any referral pressure, chose to subscribe to a niche EdTech app for a competitive exam. That is the signal that matters. It confirms the core hypothesis.
The paywall is live and generating passive exposure: 10–45% of daily active users see it during a session.
04 · Product traction — last 28 days
| Metric | Value | Context / Change |
|---|---|---|
| Total installs | 2.13k | +174 this period |
| Monthly active users | 234 | +16% |
| Daily active users | 12.6 | +15% |
| Active devices | 437 | +6% |
| Audience growth rate | 6.97% | Stable |
| Play Store conversion | 28.3% | Visitor → install |
| App Store downloads | 77 | +13.2% |
| App Store conversion | 10.9% | +88.9% |
The first full week (since PostHog product analytics went live on May 2, 2026) produced 76 weekly active users.
Activation funnel (PostHog, 82 logged-in users)
| Step | Metric |
|---|---|
| Logged in | 82 |
| Reached home screen | 91.5% |
| Started a practice set | 62.2% |
| Completed a quiz | 18.3% |
| Completed the course | 65.5% |
Course completion: 65.5% Most EdTech products celebrate anything above 30%. 19 of 29 users who entered "Understanding Verbal" completed it end-to-end, with gradual rather than sharp drop-offs. This is strong product-market signal for the course format.
Quiz completion: 18.3% Question-level data shows answers dropping from 62 at Q0 to 19 by Q9, a 70% abandonment through the quiz. Users are interested enough to start. Something about pacing or mid-quiz UX loses them. This is the primary product lever to pull next.
What sections users practice: Para-completion leads with 73 starts, followed by Para-jumbles (32), Facts/Inferences and Syllogisms (16 each). The heaviest format attracting the most sessions means serious aspirants (not casual browsers) are the ones finding the app.
Distribution — Instagram as the growth engine
With TikTok banned in India, Instagram Reels is the primary short-form content platform for this demographic. The content strategy is lean and AI-native: ChatGPT images generate avatar visuals, text overlays are composed with Claude and ChatGPT, and Veo 3 generates talking-head reels with scripted hooks. The full production loop (hook, script, visual, posting) runs with minimal manual effort.
Instagram Performance (90 days)
| Metric | Value | Context |
|---|---|---|
| Total views | 5,240 | 99.2% from non-followers |
| Accounts reached | 42,137 | 99.5% via Reels |
| Profile visits | 4,248 | From content discovery |
| Accounts engaged | 150 | 96.4% non-followers |
| Total interactions | 165 | 97.8% from Reels |
| Followers gained | 22 | |
| External link taps | 12 |
The reach-to-follower ratio tells an important story: 42,000 accounts reached with only 22 followers gained means the content is finding new audiences, not just preaching to the converted. A 28.3% Play Store visitor-to-install conversion rate suggests users who do land on the listing are intent-driven, not casual.
The current bottleneck is the gap between reach and conversion. Profile visits (4,248) versus external link taps (12) indicates the bio-to-store journey has friction. This is a solvable distribution problem, not a demand problem.
Roadmap — what comes next
- Fix mid-quiz retention. The 70% drop-off inside the quiz is the highest-leverage product fix. Likely interventions: progress indicators, shorter default quiz length, better question pacing, or a save-and-resume mechanic. This directly unlocks paywall exposure quality — only users who complete quizzes understand the value of unlimited access.
- Content expansion with AI tooling. Content production has been the constraint — limited time as a solo founder. The plan is to use AI writing tools to accelerate question set creation, explanation writing, and course content at meaningful volume, targeted for CAT 2025–26 season (exams in November–December).
- Instrument and optimise the paywall. 10–45% of DAUs see the paywall every session. With proper A/B testing, conversion copy, and pricing experiments, even a 2–3% conversion rate on 234 MAUs generates meaningful MRR. This is the most immediate revenue lever.
- Instagram distribution at scale. The AI content stack (Veo 3 + ChatGPT + Claude) is proven. Increasing posting frequency and testing hooks more aggressively is a direct path to growing the 42k reach figure and converting more of it to installs.
- Adjacent exam verticals. The same app architecture, AI content pipeline, and distribution playbook is transferable. The TMUA (Test of Mathematics for University Admission) in the UK is an early candidate — a high-stakes, niche exam with underserved preparation tooling. This is the template for a portfolio of exam-prep apps.
Summary
| Signal | Reading | Status |
|---|---|---|
| Willingness to pay | 3 subscribers, all organic, zero acquisition spend | Validated |
| Activation rate | 62% of logged-in users start a practice set | Strong |
| Course completion | 65.5% end-to-end — rare for EdTech at any stage | Exceptional |
| MAU growth | +16% month-on-month, zero paid acquisition | Healthy |
| Instagram reach | 42k accounts reached in 90 days, AI-native content stack | Working |
| Quiz completion | 18.3% funnel — 70% mid-quiz abandonment | Watch |
| Week-1 retention | 12.5% — too early to judge, needs more volume | Watch |
| Paywall conversion | High passive exposure, conversion not yet instrumented | Pending |
| Content depth | Constrained by solo founder bandwidth; AI tooling planned | Watch |
VARC for CAT is not a venture-scale bet. It is a focused, profitable-by-design product with a proven audience, a working distribution channel, and a clear content moat to build. The CAT aspirant demographic (300,000 people per year, many willing to pay for quality preparation) is large enough to build a real business on, and underserved enough that a well-executed product wins.
The infrastructure is in place: both store listings, the subscription stack, analytics, and a repeatable AI-assisted content and marketing pipeline. What scales from here is content volume and distribution intensity, both of which are solvable with more time or capital than a solo founder currently commands.
For an acquirer already operating in the Indian EdTech space, or for a portfolio looking for a proven exam-prep template to replicate across verticals (TMUA, LSAT, GMAT, JEE subsections), this is a low-risk, high-optionality asset.