सSaral Shiksha Yojna
Courses/Technology Product Entrepreneurship

Technology Product Entrepreneurship

CS9.424
Ramesh Loganathan + Prakash Yalla•Monsoon 2025-26•4 credits
Sample Papers/TPE end-sem mock paper · Paper 4

TPE end-sem mock paper · Paper 4

Duration: 120 min • Max marks: 100

Part 1 — Higher-Order Concepts (20 Marks Total, @5 each)

0 marks
  1. 1.Customer Segments for a Genomics Diagnostic A startup is building a low-cost, at-home genomic-risk-screening kit (saliva sample) that returns hereditary disease-risk insights through a mobile app. Identify: a) One specific target customer; b) One key problem they face; c) Why they may pay for this solution; d) One way to test interest quickly.5 m
  2. 2.Problem–Solution Fit for SMB Accounting A startup wants to help small business owners (5–25 employees) close their monthly books faster using AI-driven categorisation of bank transactions and GST invoices. a) Who is the customer; b) What is the main problem; c) What is one simple MVP they can test first; d) What concern might the SMB owner have that the founders may underestimate?5 m
  3. 3.Idea Hexagon — Industry 4.0 / Manufacturing Using the Idea Hexagon, generate six startup ideas in Industry 4.0 / smart manufacturing (e.g., predictive maintenance, quality inspection, factory analytics, robotic process automation on the shop floor). Each idea must clearly specify the target user and the core problem being solved.5 m
  4. 4.Business Model Basics — Sustainable Packaging A startup makes compostable food-packaging containers (cups, lids, takeaway boxes) for cloud kitchens and quick-service restaurants. Identify: a) One target customer segment; b) The core value proposition for this segment; c) One possible revenue model; d) One channel to reach customers.5 m
  5. 5.Your TPE Startup Idea (Bonus 5 Marks) Write down a one-line ad-lib for your idea as per the framework shared in class. State: a) Target user; b) Main customer pain; c) Product benefit; d) One gain for the customer. ---

Part 2 — Advanced Framework Application (40 Marks total, @20 each)

0 marks
  1. 1.Case: InspectorAI for Auto Parts InspectorAI is building a visual-AI quality-inspection system for automotive parts manufacturers. Cameras mounted on the production line detect surface defects in real time, replacing manual inspection. Quality heads are excited, but plant managers worry about line stoppages during deployment and integration with existing PLCs. A few global incumbents (Cognex, Keyence) dominate at premium prices. Questions: a) Identify the beachhead customer segment (5) b) List two customer pains and two gains the product addresses (5) c) Explain how InspectorAI can differentiate from premium global incumbents (5) d) Suggest one go-to-market strategy for early adoption (5)20 m
  2. 2.Case: StudyBuddy AI Tutor StudyBuddy is an AI tutor for grade 9–12 students preparing for engineering entrance exams. Students enjoy the personalised problem sets, but only 22% return after week 2. Parents are willing to pay, but they want measurable improvement before renewing. Existing competitors (BYJU's, PW, Unacademy) have stronger brand presence and deeper content libraries. Questions: a) Identify two hypotheses the startup must test (5) b) Classify each as a problem risk or a solution risk (5) c) Suggest one MVP experiment to improve week-2 retention (5) d) Propose one revenue model that aligns parent willingness-to-pay with measurable improvement (5) ---20 m

Part 3 — Deep-Dive Case Study (40 Marks)

0 marks
  1. 1.At-Home Genomic Screening **a) Target customer:** Health-conscious urban couples aged 30–45 planning pregnancy or with one parent who has a family history of hereditary disease — they have disposable income and an immediate decision-making context. **b) Key problem:** No accessible way today to know hereditary risk without a hospital visit, a referral chain, and ₹15K+ for clinical-grade testing. **c) Why pay:** Peace of mind during family planning is highly valued; the cost of *not* knowing (a preventable hereditary condition) is catastrophic. Anchored against clinical testing, ₹5–8K feels like a steal. **d) Quick test:** A pre-order landing page targeted at couples in 2 cities with a ₹500 refundable deposit; measure conversion, who actually deposits (couple? individual? gifted by parent?), and survey them on what risks they most fear knowing about.
  2. 2.SMB Accounting AI **a) Customer:** SMB founder or in-house accountant in a 5–25 employee business — typically an outsourced CA today handles GST filings. **b) Main problem:** Monthly book closure takes 7–12 days because transactions across multiple bank accounts and GST invoices have to be manually categorised; cashflow visibility lags reality by a full month. **c) Simple MVP:** A spreadsheet/upload tool — owner uploads one month of bank statement + one month of GST invoices; system returns categorised P&L in under 2 hours. Run with 30 SMBs. Measure: time to close vs their current baseline. **d) Underestimated concern:** SMB owners' deeper fear is *not* speed — it's **what the books reveal** to their CA, banker, or family. Many run informal practices that they don't want a tool to surface. A faster, sharper view of their books may feel like a threat. Founders must position the product as *the owner's* private intelligence tool, not a compliance broadcast.
  3. 3.Idea Hexagon — Industry 4.0 1. **Generalize:** Predictive maintenance built for CNC machines → extend to any rotating asset (pumps, compressors, motors, HVAC). Target: facility managers; problem: unplanned downtime. 2. **Fusion:** Visual inspection AI + worker safety monitoring → one camera system detects defects *and* PPE violations. Target: plant managers; problem: combined quality + safety costs. 3. **Find the Nails:** A vibration-sensor + edge-AI module → applicable to bearing health, gearbox monitoring, wind turbines, vehicle diagnostics, structural health of bridges. 4. **Find the Hammers:** Worker training problem → solutions include AR-headset guides, simulation-based training, peer-mentor matching, gamified upskilling apps. 5. **Add an Adjective:** *No-code* factory analytics — plant managers build their own dashboards by dragging shop-floor data sources, no IT team needed. Target: plant managers in mid-size factories. 6. **Do the Opposite:** Instead of pushing data from the factory floor upward to enterprise systems, push real-time decisions from a central AI brain *downward* to autonomous machine adjustments — closed-loop manufacturing optimisation.
  4. 4.Sustainable Packaging for QSRs **a) Target segment:** Mid-size cloud-kitchen brands (5–50 outlets) and city-specific QSR chains that have made public ESG commitments or face local plastic bans. **b) Value proposition:** Compostable packaging that costs only 8–12% more than plastic, ships in QSR-friendly form factors (leak-proof, microwave-safe, branded), and includes a third-party certificate of compostability that the QSR can put on its menu and Instagram. **c) Revenue model:** Per-unit pricing on a 6-month volume commitment + a small monthly subscription for branded design + a one-time setup fee. Multi-revenue, predictable. **d) Channel:** Cloud-kitchen aggregators (Curefoods, Rebel Foods, Box8) as bulk-deal partners; once 1–2 large cloud-kitchen players sign, the network effect cascades through their kitchen network.
  5. 5.Your TPE Startup Idea (Bonus) — sample *Ad-lib:* "For [smallholder farmers in tier-3 districts] who [lose 20–30% of crop value to inefficient irrigation], our [solar-powered soil-moisture sensor with SMS alerts] is a [precision-agriculture tool] that [tells them exactly when and how much to irrigate, in vernacular language]. Unlike [generic farm-management apps], our product [needs no smartphone, no internet, and pays back in one season]." **a)** Smallholder farmers in tier-3 districts. **b)** Inefficient irrigation costing 20–30% of crop yield. **c)** Vernacular SMS alerts driving precise irrigation, no smartphone needed. **d)** Higher yield per acre + lower water + power bills.
  6. 6.InspectorAI for Auto Parts **a) Beachhead:** **Tier-2 auto-component manufacturers** (~₹100–500 Cr revenue) supplying to OEMs — they have manual inspection costs, can't afford incumbent solutions, and have shorter decision cycles than tier-1 supplier giants. **b) Pains:** (i) High inspection cost — manual QC headcount of 15–40 inspectors; (ii) Defects shipped to OEMs trigger penalty clauses, sometimes terminating supply contracts. **Gains:** (i) 60–80% reduction in inspection labour cost; (ii) Verifiable defect logs that strengthen OEM negotiations and audits. **c) Differentiation from Cognex/Keyence:** - *Pricing model* — pay-per-defect-detected SaaS rather than ₹50–80 lakh upfront capex. - *Indian factory fit* — works with existing 2D cameras and older PLCs that incumbents don't bother supporting. - *Local service* — 4-hour on-site support across industrial belts (Pune, Chennai, Pithampur), whereas incumbents fly engineers from Bangalore or Singapore. - *Defensibility* — proprietary Indian-component defect dataset grows with every customer, creating a Tier-2 USP (Difficult to imitate). **d) GTM strategy:** **"30-day free pilot, pay only on results."** Offer a free 30-day deployment on one production line; the customer pays only if defect detection rate exceeds 95% (third-party audited). This removes the deployment-fear objection and creates referenceable case studies in tier-2 belts within months.
  7. 7.StudyBuddy AI Tutor **a) Two hypotheses:** - H1: Students return more often when they see measurable rank/score improvement (vs general "learning progress"). - H2: Parent willingness-to-renew depends on observable mock-test results, not on AI-tutor engagement metrics. **b) Classification:** - H1 is a **solution risk** — the AI personalisation may not produce visible enough score lift to keep students returning. - H2 is a **problem risk** — the founders may have misidentified the customer's true problem (parents don't care about engagement metrics; they care about JEE/NEET ranks). **c) MVP experiment for week-2 retention:** Split 500 new users into three cells. *Cell A:* current product. *Cell B:* every session ends with a 10-question mock-test snapshot showing rank percentile change. *Cell C:* same as B + a peer leaderboard with friends from the same coaching centre. Measure week-2 return rate, session count, and parent-reported renewal intent. **d) Revenue model:** **Outcome-linked pricing.** Parents pay ₹2,000 base + ₹3,000 if mock-test rank improves by 10 percentile within 90 days (audited via partner-conducted tests). This aligns the company's incentives with parent willingness-to-pay, generates referrals (parents brag about results), and forces product focus on score lift rather than vanity engagement. ### Part 3 — LoanMint MSME Credit **a) Strategic Evaluation (15 marks)** | Option | PMF | Revenue potential | Sales complexity | Scalability | |---|---|---|---|---| | Direct NBFC (own balance sheet) | Strong — tier-2 retailers respond, 28% approval shows demand | **Very high** at scale — net interest margin of 8–14% on ₹100Cr+ AUM | **Low** for borrowers (B2C-style) but **very high** regulatory complexity | **Low–Moderate** — capital-intensive; growth gated by balance sheet | | Lending marketplace | Strong — borrowers want speed, lenders want volume | Moderate — referral/origination fees of 1.5–3% per loan | **Moderate** — multi-sided cycle (lenders + borrowers) | **High** — capital-light, scales with platform reach | | SaaS licensing (underwriting engine) | Unproven scale — only 3 enterprise leads so far but interest is real | **High** per deal (₹50L–₹2Cr/yr) but slow to compound | **Very high** — 6–12 month enterprise cycles, custom integration | **High** — pure tech, classic SaaS scale economics | **b) Recommendation (10 marks)** With 15 months of runway, **operate as a lending marketplace as the primary path**, while parallel-running early SaaS-

Track your attempt locally — score and time are recorded in your browser. (Coming soon: timed-attempt mode.)