SB StartupBasket
All ideas
76 /100 GO Low complexity

ClaimVault — topper-ad proof desk for coaching institutes

Proves which course every topper in your ad actually bought, before the CCPA asks and fines you.

EdTech SaaS India Compliance-drivenSMBWhatsApp-firstAI-agentMultilingual
— views
Evaluation Scores
76/100

GO

Overall Score

16
Problem
13
Demand
13
Build
12
Distrib.
11
Revenue
8
Time
3
Defense

ClaimVault

1. One-liner

Proves which course every topper in your ad actually bought, before the CCPA asks and fines you.

2. Trend signal — why now?

The Central Consumer Protection Authority stopped issuing advisories and started issuing cheques-payable-to-government.

The Guidelines for Prevention of Misleading Advertisement in Coaching Sector, 2024 took effect 13 November 2024. They apply to “all forms of advertisement by any person who provides coaching or establishes, runs or administers a coaching center, including endorsers.” The core obligation is not “don’t lie” — it is disclose the material information, specifically the course each successful candidate actually took, displayed with the same prominence as the claim. Disclaimer and claim cannot be in different fonts.

Then enforcement arrived, and it is escalating on a visible curve:

  • 45 notices, ₹61.6 lakh on 19 institutes — early enforcement tranche.
  • December 2025 — Vision IAS (AjayVision Education Pvt Ltd), ₹11 lakh. The institute claimed credit for 119+ successful candidates across UPSC CSE 2022 and 2023. CCPA found only 3 had enrolled in foundation courses; the other 116 bought Test Series or Mock Interview programmes. CCPA Chief Commissioner Nidhi Khare: “This is the first case of penalty on the second offence.”
  • February 2026 — Vajirao & Reddy Institute, ₹15 lakh.
  • May 2026 — Motion Education ₹10 lakh, Career Line Coaching Sikar ₹5 lakh; Vajiram and Ravi ₹7 lakh — the last for advertising top-10 rankers where 7 of 8 had only taken the free Interview Guidance Programme.
  • Cumulative: 60+ notices, penalties exceeding ₹1.46 crore.

Meanwhile ASCI’s Half-Yearly Complaints Report 2025–26 puts education as the most non-compliant category, topping violation charts again, with 71 education ads investigated and 45% voluntarily withdrawn after intervention. 97% of all violations were digital ads — the exact channel where a coaching centre ships creatives fastest and reviews them least.

The pattern in every single penalty order is identical and mechanical: the institute could not evidence which product each named rank holder bought. That is a record-keeping failure, not a creative failure. Record-keeping failures are software-shaped.

Provenance:

3. The opportunity

Every coaching institute in India runs the same annual machine. Results drop in May–June. Within 48 hours the marketing team must ship topper posters, Instagram reels, WhatsApp broadcasts, hoardings, newspaper jackets and paid creatives — hundreds of assets, produced by a junior designer or a ₹25,000/month agency, approved by a director on his phone between admission calls.

Nobody in that chain has access to the CRM record showing that Rank 47 bought a ₹9,500 test series, not the ₹1.85 lakh foundation course. The designer has a photo, a name and a rank. So the creative says “Our Student — AIR 47” and the institute has just committed the exact violation that cost Vision IAS ₹11 lakh.

The incumbent alternatives are all wrong-shaped:

  • Law firms will audit your campaign for ₹50K–₹2L per engagement, take two weeks, and be useless at result-season tempo when you need 300 creatives cleared in 72 hours.
  • Ad agencies produce the offending creative. They are the risk, not the control.
  • Admission CRMs (LeadSquared, ExtraaEdge, HamaraCRM, Teachmint) hold the enrolment truth but have zero concept of an advertising claim. They will never build this — their buyer is the admissions head, not the compliance-exposed director.
  • ASCI pre-screening is advisory and voluntary. It does not produce the evidence file CCPA demands.

The gap is a system of record that sits between the enrolment data and the creative, and can answer one question on demand: for this specific ad, what did each named student actually buy, and can I prove it in a file if a notice lands?

This is a disruption play against manual legal review, and the AI does real work: reading unstructured creatives in Hindi and English, extracting claims, and matching named students against enrolment records.

4. Target market

  • Primary customer: Founder/director or marketing head of a coaching institute with ₹2Cr–₹200Cr revenue — UPSC, JEE, NEET, CA, banking, state PSC. 1 to 40 centres. Kota, Delhi (Mukherjee Nagar/Karol Bagh), Jaipur, Sikar, Hyderabad, Patna, Indore, Pune. The buyer is the person whose name is on the CCPA order.
  • Why they buy: Not fear of ₹250 crore — that is vendor theatre, and this buyer is too streetwise for it. They buy because penalties in their actual band are ₹5–15 lakh and rising, because CCPA now punishes repeat offences harder, because orders are covered by name in national press during admission season, and because reputational damage in a word-of-mouth market costs more than the fine. Motion Education and Vajiram and Ravi are household names in this sector; every director read those stories.
  • Rough TAM reasoning: The hardest number available is the government’s own: GST collected from coaching institutions rose from ₹2,240 crore (FY2019-20) to ₹5,517 crore (FY2023-24) — 146% growth in five years (Ministry of Education, Rajya Sabha Unstarred Q. No. 983, 31.07.2024). That figure captures only the tax-registered, organised tier, which is precisely the set that advertises results publicly and has revenue worth protecting. Physics Wallah’s RHP puts test-prep at ₹1–1.1 trillion in FY25, ~13% CAGR — a regulatory filing, not a vendor estimate. There is no credible national count of coaching centres; the same Parliament answer declined to give one, referring only to “unregulated private coaching centers… in the absence of any laid down policy,” and I am not going to invent one. Work bottom-up instead: CCPA has already noticed 60+ institutes. Conservatively 8,000–15,000 institutes nationally run enough paid advertising to need this. At 1,200 customers this is a ₹4Cr+ ARR business, which is the whole point.
  • Why now for them: The 2026 result season just demonstrated that CCPA is fining mid-size institutes, not just Delhi giants. Career Line Coaching in Sikar is not a national brand and still got ₹5 lakh. Regulatory pressure is also converging from a second direction: the Rajasthan Coaching Centres (Control and Regulation) Act, 2025 (Act 16 of 2025, passed 3 Sep 2025) makes registration mandatory for centres above 100 students, requires each branch to register separately, and escalates to ₹2 lakh penalties and cancellation. Assam approved a similar bill. A director in Kota or Sikar is now being formalised and inspected from two sides at once. The next result cycle is April–June 2027 — that is the buying window.

5. Product sketch (MVP)

  • Creative intake — drop a poster, reel, PDF, or WhatsApp broadcast image into the dashboard or forward it to a WhatsApp number. Handles Hindi and English, print and digital.
  • Claim extraction — automatically pulls every factual assertion out of the creative: named students, ranks, selection counts, percentages, guarantee language, urgency/scarcity claims, fee and refund promises.
  • Substantiation match — checks each named student against the institute’s uploaded enrolment records (CSV export from their CRM or fee software) and returns the course actually purchased, duration, and fee paid.
  • Red/amber/green verdict per creative — red where a claimed topper bought a different product than the ad implies, amber where the required disclosure is missing or under-prominent, green where the claim is substantiated and disclosed.
  • Disclosure text generator — produces the compliant equal-prominence line (“AIR 47 — enrolled in Test Series 2025, 4 months”) sized to the creative, in the ad’s language.
  • Consent tracker — logs written post-selection consent per candidate for name/photo/video use, with expiry and renewal reminders. This is a separate guideline obligation most institutes are quietly failing.
  • Evidence pack export — one-click PDF per campaign: every creative, every claim, the matching enrolment record, consent proof, and timestamps. This is the file you hand a CCPA notice.
  • Repeat-offence register — tracks which claims were previously flagged or withdrawn, so the same violation never ships twice. Directly targets the escalation that cost Vision IAS ₹11 lakh.

6. AI angle — what’s load-bearing

Remove the AI and this is a spreadsheet nobody fills in — which is precisely the status quo that keeps generating penalties.

The load-bearing work is claim extraction from messy multilingual visual creatives. A topper poster is a JPEG with Devanagari and English text baked into it at odd angles over a photograph, produced in Canva by a 23-year-old. Getting from that to a structured list of assertions — “claims AIR 47; implies foundation course enrolment; no course disclosure present; urgency claim ‘only 20 seats left’” — is a vision-plus-language task that became cheap and reliable in the last 18 months. Before that you needed a human reviewer per creative, which is exactly why institutes shipping 200+ creatives a year do not review them.

Second load-bearing piece: fuzzy entity matching between the name printed on a poster and the name in a fee ledger, across transliteration variants, nicknames, and roll numbers. Rule-based matching fails on Indian name data; this is where most of the accuracy work lives.

7. Localization angle

This is India-only by construction — it exists because of an Indian regulator’s guidelines and an Indian market’s topper-advertising culture.

  • Language: Creatives are Hindi, English, and increasingly regional (Marathi, Telugu, Tamil, Bengali). Extraction must work on baked-in Devanagari text, not just Unicode.
  • Distribution: WhatsApp is the primary channel — both for how creatives circulate and for how the product should be operated. Forwarding a poster to a WhatsApp number and getting a red/amber/green reply in 90 seconds is the right interface for a marketing head during result week, not a web dashboard.
  • Pricing: ₹6,000–₹25,000/month works for this wallet against a ₹5–15 lakh penalty and ₹50K+ per legal opinion. A $49/mo global compliance SaaS has no equivalent here because no other country regulates coaching topper ads this specifically.
  • Payment rails: UPI autopay and annual prepay — this buyer prefers a single annual invoice tied to the admission cycle.

8. Business model — path to $1M–$5M ARR

  • Pricing: Three tiers. Starter ₹6,000/mo (single centre, 50 creatives/mo). Growth ₹15,000/mo (multi-centre, 250 creatives/mo, consent tracker). Institute ₹25,000/mo (unlimited creatives, multi-brand, evidence-pack API, priority result-season SLA). Annual prepay at 10 months’ price — matches how this sector buys.
  • ACV: ₹1.4 lakh blended (~$1,650).
  • Rough math to $1M ARR: 600 institutes × ₹15,000/mo × 12 = ₹10.8Cr ≈ $1.27M. 600 customers out of a conservative 8,000–15,000 advertising institutes is 4–7% penetration.
  • Rough math to $5M ARR: ~2,500 customers at a slightly higher blended ACV, plus expansion into adjacent regulated advertisers with the same claim-substantiation problem — private universities, skill-development institutes, and study-abroad consultancies, all of which advertise placement and selection outcomes under the same Consumer Protection Act exposure.
  • Expansion path: Per-creative volume as institutes grow; per-brand for groups running multiple exam verticals; a done-for-you review service during the April–June crunch at ₹50K–₹1L per season, which is high-margin and sells itself when the marketing team is drowning.

Gross margin is strong — inference on a few hundred images a month is rupees, not thousands of rupees.

9. Go-to-market wedge — first 100 customers

  1. The penalty list is a lead list. CCPA orders name the institute. 60+ have been noticed publicly; every one is a warm prospect who has personally felt this. Pull the named institutes from CCPA press releases, PIB, and national coverage, find the director on LinkedIn or via the institute’s own contact page, and send a one-page teardown of their own published creative showing exactly which claims lack substantiation. Not a demo — their actual poster, marked up. Expect strong reply rates from people holding a live notice; target 15 of the first 100 here.
  2. Result-season ambush, April–June 2027. Scrape topper creatives from the Instagram and Facebook accounts of 500 institutes in Kota, Sikar, Jaipur, Delhi, Patna, Hyderabad. Auto-generate a free risk report per institute. Send it over WhatsApp within 72 hours of their post going live, while the fear is fresh and the creative is still running. This is the highest-intent moment in the entire calendar; 40 of the first 100 come from here.
  3. Coaching-federation and local-association route. State coaching associations exist in Rajasthan, Bihar, UP and Telangana, and members share regulatory scare stories constantly. Offer a free 45-minute session on “what the CCPA orders actually require” to association meetings — a genuinely useful teardown of the Vision IAS and Vajiram orders. Convert 10–20% of the room. 25 of the first 100.
  4. The agencies that make the ads. A few hundred small agencies serve this sector and are terrified of being blamed. White-label the checker to them at ₹10,000/mo so they can sell “CCPA-cleared creatives.” Each agency brings 5–20 institutes. 20 of the first 100.

Every one of these channels reaches a named, findable buyer with a documented problem. None of them is “content marketing.”

10. Build complexity — justification

Low. Off-the-shelf multimodal models handle creative ingestion and claim extraction; the enrolment match is a CSV import plus fuzzy name matching; the evidence pack is PDF generation. No integrations are strictly required for v1 — institutes export from their CRM or fee software and upload, which is how they already work. WhatsApp Business API for intake is standard.

The genuine engineering is accuracy tuning on Indian name matching and baked-in Devanagari extraction, plus building the guideline rule-set correctly. A pair ships a credible v1 in 8–10 weeks. The hard part of this business is domain interpretation and sales access, not code.

11. Gating checklist

GatePass?Note
Legal in target market✅Helps institutes comply with published CCPA guidelines. No grey area.
Ethical — no harm / dark patterns✅Genuinely protects students and parents from misleading claims. Refuses to help disguise a false claim — it flags it.
Market exists (evidence above)✅₹1.46 crore in levied penalties, 60+ notices, named institutes, ASCI’s most-violating category.
1–5 person team can build this✅Two people, 8–10 weeks.
Launchable with <$50K / ₹40L✅Well under ₹15L including a domain expert on retainer.

All five pass.

12. Feasibility score

AxisWeightScoreNotes
Problem intensity2016/20Real money already leaving real institutes: ₹5L–₹15L per order, escalating for repeat offences, named in national press during admission season. Not hair-on-fire year-round though — intensity spikes hard around results and notices, and dips between.
Demand evidence1513/15Unusually hard evidence: 60+ notices, ₹1.46Cr levied, named institutes across 2025–2026, ASCI ranking education the most non-compliant category. Docked 2 because I found no verbatim operator complaints — the evidence is regulatory and press, not forum voice.
Build feasibility1513/15Off-the-shelf multimodal extraction plus CSV matching. Pair ships in 8–10 weeks. Accuracy tuning on Indian names is the only real engineering risk.
Distribution clarity1512/15The penalty list is a public, named lead list, and result-season ambush is a precise high-intent moment. Docked because this buyer is offline-first and needs phone/relationship selling — slower than pure self-serve.
Revenue mechanics1511/15₹6K–₹25K/mo is well-benchmarked against ₹50K legal opinions and lakh-scale penalties. Docked because seasonality is real: churn risk in the Aug–Feb trough is the central unknown.
Time to first revenue108/10Institutes holding live notices will pre-pay quickly; 6–8 weeks to first cheque is realistic. Not 4 weeks, because this is a relationship sale.
Defensibility103/10Honestly weak. Execution-and-focus moat only. The guideline rule-set is public and copyable; a CRM incumbent could bolt this on if the category proves out. Real defence is accumulated claim-precedent data and being the name directors mention to each other.
Total10076/100

13. Qualitative modifiers

Founder-fit tags

sales-heavy · domain-expertise-required

This is not a build-it-and-they-come product. It needs somebody who will sit in a Kota director’s office, and somebody who can read a CCPA order correctly. The engineering is the easy half.

Key assumptions to validate (3–5)

  1. Assumption: Institutes will pay ₹15,000/mo for pre-publication clearance rather than absorb an occasional ₹5–15 lakh fine as a cost of doing business. How to test: Take the 25 most recently noticed institutes, present a marked-up teardown of their own creative, and ask for a ₹25,000 annual prepay commitment before building anything. Need 5+ commitments.
  2. Assumption: Institutes can produce usable enrolment records mapping a named topper to a purchased course. If their own data is a mess, the product cannot substantiate anything. How to test: Ask 10 institutes for a sample enrolment export and attempt the match manually against a real published poster of theirs.
  3. Assumption: Claim extraction from Hindi/English baked-text creatives is accurate enough to trust. How to test: Run 200 real scraped topper creatives; measure precision/recall on claim identification against manual review. Need >90% recall on named-student claims.
  4. Assumption: Demand is not fatally seasonal — institutes retain subscriptions outside April–June. How to test: Structure early deals as annual prepay and track renewal intent at month 6; alternatively test a seasonal-service price point.

Risk flags

  1. Regulatory dependency: The entire product exists because CCPA keeps enforcing. If enforcement goes quiet for 18 months, urgency evaporates. Mitigate by expanding into the broader Consumer Protection Act claim-substantiation problem across adjacent education advertisers.
  2. Seasonality: Revenue pressure and usage concentrate in April–June. Annual prepay contracts are essential; monthly billing would expose brutal trough churn.
  3. Low defensibility: An admission-CRM incumbent already holds the enrolment data and could add claim-checking. Speed and a compliance-specific brand are the only answers; assume 12 months of clear air, not 36.
  4. Buyer moral hazard: Some institutes will want a tool that helps them dress up a claim rather than substantiate it. Refusing that business is both the ethical and the correct commercial call — the product’s value is that its evidence pack stands up to a regulator.

14. Structured verdict

Score:                  76/100
Verdict:                GO
Confidence:             Medium
Best-fit builder:       Sales-led operator with India edtech/coaching-sector access, paired with one engineer; needs a consumer-law advisor on retainer
Time to revenue:        6–8 weeks
Capital to launch:      ₹10–15 lakh ($12–18K)
Top 3 assumptions to validate first:
  1. Willingness to pay ₹15K/mo — pitch 25 recently-noticed institutes with a teardown of their own creative; need 5 prepay commitments
  2. Enrolment data is matchable — collect 10 sample exports, attempt manual substantiation against their real published posters
  3. Extraction accuracy on Hindi/English baked-text creatives — 200-creative benchmark, need >90% recall on named-student claims
Kill criteria:
  - Abandon if fewer than 3 of 25 noticed institutes will commit money before build
  - Abandon if >40% of sampled institutes cannot produce enrolment records that map a named topper to a purchased course
  - Abandon if CCPA issues no new coaching penalties through the entire April–June 2027 result season

15. Next step — 1-week validation sprint

  • Day 1–2: Pull every publicly named CCPA coaching penalty and notice from PIB, CCPA releases and press coverage. Build a list of 40 institutes with director contact details. In parallel, scrape 200 live topper creatives from those institutes’ Instagram and Facebook pages.
  • Day 3–4: Hand-produce marked-up teardowns for 25 of them — their own creative, every unsubstantiated claim circled, the missing equal-prominence disclosure named, referencing the specific guideline clause and the Vision IAS repeat-offence order. This is manual, and that is the point: if I cannot do it by hand, no model will.
  • Day 5: Send all 25 over WhatsApp and email. Ask for one thing: ₹25,000 annual prepay for the 2027 result season, refundable.
  • Decide go / no-go on: ≥5 of 25 prepay commitments and ≥6 institutes willing to share an enrolment export. Below that, the fine is being treated as a cost of doing business and I walk.

Falsifiable, cheap, and it produces money or a clear no within seven days.

Interested in a detailed proposal?

Get a deep-dive with market research, competitive analysis, and implementation roadmap.

Contact us

info@startupbasket.ai