Analytical Skills-II

Unit 2: Syllogism, Number Ranking & Guesstimation

Master logical deductions, pattern recognition, and structured estimation — the three pillars of analytical thinking tested across Bank PO, CAT, placements, and consulting interviews.

⏱️ 6 hrs theory + 4 hrs practice | 🎯 Bank PO / CAT / Placement | 💰 Consulting Essential

💼 Careers this unlocks: Management Consultant (₹12–25 LPA) | Bank PO (₹6–10 LPA) | Business Analyst (₹5–9 LPA)

Section A

Opening Hook — The Question That Gets You Hired

🏢 "How Many ATMs Are There in India?" — Answer This Correctly and You're Hired

McKinsey, BCG, Bain — the top 3 management consulting firms in the world — ask guesstimation questions in EVERY single interview. Not because they want the exact number. They want to see how you think.

A candidate who says "I don't know, maybe 2 lakh?" gets rejected. A candidate who says "Let me break this down — India has ~1.4 billion people, approximately 150 crore bank accounts, the RBI reported ~2.15 lakh ATMs as of 2024, and if I estimate using bank branches..." gets hired.

But guesstimation is just ONE tool in your analytical toolkit. This chapter also covers Syllogism — the backbone of logical reasoning in Bank PO, SSC, and CAT exams — and Number Ranking Tests, which appear in every placement aptitude round.

Master these three, and you dominate every competitive exam AND every consulting interview.

🏢 McKinsey🏢 BCG🏢 Bain🇮🇳 SBI PO🇮🇳 IBPS🎓 CAT/XAT

McKinsey India alone receives 50,000+ applications per year and hires only ~200. Every single shortlisted candidate faces at least 2 guesstimation questions. The skill you learn in this chapter literally separates the top 0.4% from the rest. BCG's Mumbai and Gurgaon offices use identical interview formats.

Section B

Learning Outcomes — Bloom's Taxonomy Mapped

Bloom's Level	Learning Outcome
🔵 Remember	Recall the 4 standard syllogistic propositions (All, Some, No, Some Not) and their Venn diagram representations
🔵 Remember	List the rules for combining syllogistic premises (All+All, All+Some, Some+Some) and their conclusions
🟢 Understand	Explain why "All A are B" does NOT mean "All B are A" using real-world Indian examples
🟢 Understand	Describe the difference between top-down and bottom-up approaches in guesstimation
🟡 Apply	Solve 3-statement syllogism problems using Venn diagrams and identify valid conclusions
🟡 Apply	Apply ranking formulas to determine position, count, and relative ordering in linear arrangements
🟠 Analyse	Distinguish between "must be true", "can be true", and "definitely false" conclusions in possibility-based syllogism
🟠 Analyse	Break down a guesstimation problem into population, segmentation, usage rate, and frequency components
🔴 Evaluate	Critique a given guesstimation answer for logical gaps, unreasonable assumptions, and missing segments
🔴 Evaluate	Assess whether syllogistic conclusions follow necessarily vs. possibly from given premises
🟣 Create	Construct original guesstimation frameworks for novel Indian market-sizing questions (e.g., "How many wedding photographers in Mumbai?")
🟣 Create	Design multi-step syllogism problems with traps and solve them for peers

Section C

Concept Explanation — Syllogism, Ranking & Guesstimation

PART I — SYLLOGISM

1. Logical Venn Diagrams — The 4 Standard Propositions

Syllogism is the art of drawing logically valid conclusions from given statements (premises). Before we solve problems, we must understand the 4 types of statements that form the building blocks of ALL syllogism questions.

Syllogism = Aadhaar verification logic. "All Indian citizens SHOULD have Aadhaar" — but does that mean "All Aadhaar holders are citizens?" NO! NRIs and even some foreign workers have Aadhaar numbers. This is exactly the kind of logical trap syllogism tests.

📐 The 4 Standard Propositions

Type A — Universal Affirmative: "All A are B"

Every member of group A is also a member of group B. A is completely inside B.

┌─────────────────────┐ │ B │ │ ┌──────────┐ │ │ │ A │ │ │ │ │ │ │ └──────────┘ │ │ │ └─────────────────────┘ "All A are B" → A ⊂ B (A is a subset of B)

Example: "All IIT students are engineering students." ✅ Every IIT student studies engineering.

Type E — Universal Negative: "No A are B"

No member of group A is a member of group B. The circles are completely separate.

┌──────────┐ ┌──────────┐ │ A │ │ B │ │ │ │ │ │ │ │ │ └──────────┘ └──────────┘ "No A are B" → A ∩ B = ∅ (no overlap)

Example: "No cats are dogs." ✅ A cat can never be a dog.

Type I — Particular Affirmative: "Some A are B"

At least one member of group A is also a member of group B. The circles overlap partially.

┌──────────┐ │ A ┌─┼────────┐ │ │ │ │ │ │ │ B │ └────────┼─┘ │ └──────────┘ "Some A are B" → A ∩ B ≠ ∅ (some overlap)

Example: "Some doctors are women." ✅ At least some doctors are female.

Type O — Particular Negative: "Some A are not B"

At least one member of group A is NOT a member of group B. Part of A lies outside B.

┌──────────┐ │ A ┌─┼────────┐ │ (this │ │ │ │ part) │ │ B │ └────────┼─┘ │ └──────────┘ "Some A are not B" → part of A is outside B

Example: "Some Indians are not Hindi speakers." ✅ Many South Indians speak Tamil, Telugu, etc.

Type	Statement	Code	Relationship	Example
A	All A are B	Universal Affirmative	A ⊂ B	All roses are flowers
E	No A are B	Universal Negative	A ∩ B = ∅	No fish are birds
I	Some A are B	Particular Affirmative	A ∩ B ≠ ∅	Some students are toppers
O	Some A are not B	Particular Negative	Part of A outside B	Some fruits are not sweet

2. Relating Groups Diagrammatically — How to Draw

The secret to solving syllogism without errors is to always draw the Venn diagram. Here's exactly how:

✏️ Step-by-Step Drawing Rules

Rule 1: "All A are B" → Draw A completely inside B

A is a smaller circle inside the bigger circle B. Note: B can have elements that are NOT in A.

Rule 2: "Some A are B" → Draw A and B overlapping

The two circles intersect. The overlapping region contains elements common to both.

Rule 3: "No A are B" → Draw A and B completely separate

Two circles with no contact at all. Zero common elements.

Rule 4: "Some A are not B" → Draw A with part outside B

Similar to the "Some" diagram, but emphasise that part of A does NOT touch B.

GOLDEN RULE: Always draw the minimum diagram — the diagram that satisfies ALL given premises simultaneously. Don't assume more than what is stated.

"All A are B" does NOT mean "All B are A"! This is the #1 mistake in syllogism. "All dogs are animals" is true, but "All animals are dogs" is false. Always remember: the subject (A) goes INSIDE the predicate (B), not the other way around.

3. Possibility-Based Questions — The Decision Framework

Modern exams (especially Bank PO and CAT) love possibility-based questions. These ask "Can this conclusion be true?" rather than "Must this conclusion be true?" Here's how to handle them:

🎯 Three-Level Decision Framework

Level 1 — "Definitely True" (Must be True)

The conclusion follows necessarily from the premises in every possible Venn diagram. No matter how you draw it, the conclusion holds.

Example: Premises: "All A are B" + "All B are C" → Conclusion: "All A are C" ✅ DEFINITELY TRUE

Level 2 — "Can Be True" (Possibly True)

The conclusion doesn't necessarily follow, but there EXISTS at least one valid Venn diagram where it holds true. It's not contradicted by the premises.

Example: Premises: "All A are B" → Can "All B are A" be true? YES — if A and B are the same set. ✅ POSSIBLY TRUE

Level 3 — "Definitely False" (Cannot Be True)

The conclusion is contradicted by the premises in EVERY possible diagram. There's no way to draw it without violating a premise.

Example: Premises: "No A are B" → Can "All A are B" be true? NO — direct contradiction. ❌ DEFINITELY FALSE

Quick test for "possibility" questions: If you can draw even ONE valid Venn diagram where the conclusion holds AND all premises are satisfied, then the answer is "CAN be true." You don't need it to work in ALL diagrams — just one.

4. Rules for Combining Premises — The Conclusion Table

When two premises share a common (middle) term, we can derive a conclusion. The table below tells you what conclusion follows from which combination:

Premise 1	Premise 2	Conclusion	Example
All A are B	All B are C	All A are C ✅	All dogs are animals + All animals are living beings → All dogs are living beings
All A are B	No B are C	No A are C ✅	All cats are pets + No pets are wild → No cats are wild
All A are B	Some B are C	No valid conclusion ❌	All roses are flowers + Some flowers are red → ??? (can't conclude about roses and red)
Some A are B	All B are C	Some A are C ✅	Some students are athletes + All athletes are fit → Some students are fit
Some A are B	Some B are C	No valid conclusion ❌	Some doctors are women + Some women are teachers → ??? (no definite conclusion)
Some A are B	No B are C	Some A are not C ✅	Some Indians are engineers + No engineers are artists → Some Indians are not artists
No A are B	All B are C	Some C are not A ✅	No cats are dogs + All dogs are animals → Some animals are not cats
No A are B	No B are C	No valid conclusion ❌	No apples are oranges + No oranges are mangoes → ??? (no definite link)

The biggest trap: All + Some = No Conclusion. Students assume "All A are B" + "Some B are C" must give a conclusion. It doesn't! The "Some B" that are C might be the part of B that is OUTSIDE A. Draw the Venn diagram to see why.

Memory trick — "AENI": When both premises are Affirmative, the conclusion (if any) is Affirmative. When one premise is Negative, the conclusion (if any) is Negative. When one premise is Particular (I/O), the conclusion (if any) is Particular. And if BOTH premises are Particular — No conclusion!

5. Worked Examples — Syllogism (10 Problems with Venn Diagrams)

Example 1: Basic Two-Premise (All + All)

Premises:

1. All mangoes are fruits.

2. All fruits are food.

Venn Diagram:

┌─────────────────────────────┐ │ FOOD │ │ ┌──────────────────┐ │ │ │ FRUITS │ │ │ │ ┌───────────┐ │ │ │ │ │ MANGOES │ │ │ │ │ └───────────┘ │ │ │ └──────────────────┘ │ └─────────────────────────────┘

Valid Conclusions:

✅ All mangoes are food. (Mangoes inside Fruits inside Food)

✅ Some fruits are mangoes. (Converse of All mangoes are fruits)

✅ Some food is fruit. (Converse)

❌ All food is mangoes. (WRONG — food contains much more than mangoes)

Example 2: All + No

Premises:

1. All teachers are educated.

2. No educated person is ignorant.

┌──────────────┐ ┌──────────────┐ │ EDUCATED │ │ IGNORANT │ │ ┌──────────┐ │ │ │ │ │ TEACHERS │ │ │ │ │ └──────────┘ │ │ │ └──────────────┘ └──────────────┘

Valid Conclusions:

✅ No teachers are ignorant. (Teachers ⊂ Educated, Educated ∩ Ignorant = ∅)

✅ No ignorant person is a teacher.

❌ All educated people are teachers. (WRONG — not all educated = teachers)

Example 3: Some + All

Premises:

1. Some politicians are honest.

2. All honest people are respected.

┌────────────────────────────────┐ │ RESPECTED │ │ ┌─────────────────┐ │ │ │ HONEST │ │ │ ┌─┼──────┐ │ │ │ │ │ (overlap) │ │ │ │ └──────┼──────────┘ │ │ │POLITI- │ │ │ │ CIANS │ │ │ └────────┘ │ └────────────────────────────────┘

Valid Conclusions:

✅ Some politicians are respected. (The honest politicians are in the respected circle)

❌ All politicians are respected. (Only SOME are honest → respected)

❌ All respected people are politicians. (Wrong direction)

Example 4: Some + Some = No Conclusion

Premises:

1. Some engineers are gamers.

2. Some gamers are musicians.

┌─────────┐ │ENGINEERS│ │ ┌────┼────────┐ │ │ │ GAMERS │ └────┼────┘ ┌────┼─────────┐ │ │ │MUSICIANS│ └────────┼────┘ │ └──────────────┘

Valid Conclusions:

❌ No definite conclusion about engineers and musicians.

The "some engineers who are gamers" may or may not be the "some gamers who are musicians." We simply cannot tell. Some + Some = No Conclusion.

Example 5: Possibility Question

Premises:

1. All pens are stationery.

2. Some stationery is expensive.

Question: Can "All pens are expensive" possibly be true?

Solution: Draw two diagrams:

Diagram 1 (where it's NOT true): Diagram 2 (where it IS true): ┌────────────────────┐ ┌────────────────────┐ │ STATIONERY │ │ STATIONERY │ │ ┌──────┐ ┌──────┐ │ │ ┌──────────────┐ │ │ │ PENS │ │EXPEN-│ │ │ │ EXPENSIVE │ │ │ │ │ │SIVE │ │ │ │ ┌──────────┐ │ │ │ └──────┘ └──────┘ │ │ │ │ PENS │ │ │ └────────────────────┘ │ │ └──────────┘ │ │ │ └──────────────┘ │ └────────────────────┘

Answer: ✅ YES, "All pens are expensive" CAN be true. Diagram 2 shows a valid scenario where all pens fall inside the expensive region. Since the premises don't PREVENT this, it's a possibility.

Example 6: Three-Premise Chain

Premises:

1. All dancers are artists.

2. All artists are creative.

3. Some creative people are wealthy.

┌───────────────────────────────────┐ │ CREATIVE │ │ ┌──────────────────┐ ┌──────┐ │ │ │ ARTISTS │ │WEALT-│ │ │ │ ┌───────────┐ │ │HY │ │ │ │ │ DANCERS │ │ │ │ │ │ │ └───────────┘ │ └──────┘ │ │ └──────────────────┘ │ └───────────────────────────────────┘

Valid Conclusions:

✅ All dancers are creative.

✅ Some creative people are artists.

❌ Some dancers are wealthy. (Cannot conclude — the wealthy overlap may not touch dancers)

✅ "Some dancers are wealthy" CAN be true (possibility). ✅ "No dancers are wealthy" CAN also be true.

Example 7: No + All

Premises:

1. No birds are reptiles.

2. All snakes are reptiles.

┌──────────┐ ┌──────────────┐ │ BIRDS │ │ REPTILES │ │ │ │ ┌────────┐ │ │ │ │ │ SNAKES │ │ │ │ │ └────────┘ │ └──────────┘ └──────────────┘

Valid Conclusions:

✅ No snakes are birds. (Snakes ⊂ Reptiles, Birds ∩ Reptiles = ∅)

✅ No birds are snakes.

✅ Some reptiles are not birds. (The snake part of reptiles is definitely not birds)

Example 8: Some + No

Premises:

1. Some students are athletes.

2. No athletes are lazy.

┌──────────┐ │STUDENTS │ │ ┌─────┼──────────┐ ┌──────────┐ │ │ │ ATHLETES │ │ LAZY │ │ │ │ │ │ │ │ └─────┼──────────┘ └──────────┘ └──────────┘

Valid Conclusions:

✅ Some students are not lazy. (The students who ARE athletes can't be lazy)

❌ No students are lazy. (WRONG — we only know SOME students are athletes; other students might be lazy)

Example 9: Bank PO Style — Complementary Pair

Premises:

1. All books are knowledge.

2. Some knowledge is power.

Conclusions to evaluate:

I. Some books are power.

II. Some books are not power.

Analysis: From "All books are knowledge" + "Some knowledge is power", we get NO definite conclusion about books and power (All + Some = No Conclusion). HOWEVER, one of the two conclusions MUST be true — either some books ARE power, or some books are NOT power. This is a complementary pair. The answer is "Either I or II follows."

Example 10: CAT Level — Four Statements

Premises:

1. All IITians are smart.

2. Some smart people are rich.

3. No rich person is unhappy.

4. All unhappy people are stressed.

Evaluate:

A) "Some IITians are rich" → ❌ Not necessarily true (All + Some = No Conclusion)

B) "Some smart people are not unhappy" → ✅ TRUE (Some smart = rich, No rich = unhappy → Some smart ≠ unhappy)

C) "No IITians are stressed" → ❌ Cannot conclude (the link from IITians to stressed is broken)

D) "Some rich people are not stressed" → ✅ TRUE (No rich = unhappy, so rich people are NOT unhappy, and only unhappy → stressed, so rich need not be stressed)

PART II — NUMBER RANKING TEST

6. Number Test — Series, Missing Numbers & Odd One Out

Number tests evaluate your ability to recognise patterns in sequences. These appear in every Bank PO, SSC, CAT, and placement exam. Here are the 10 most common types:

🔢 10 Types of Number Series

Type 1 — Arithmetic Series (Constant Difference):

3, 7, 11, 15, ? → Difference = +4 → Answer: 19

Type 2 — Geometric Series (Constant Ratio):

2, 6, 18, 54, ? → Ratio = ×3 → Answer: 162

Type 3 — Square Series:

1, 4, 9, 16, 25, ? → Pattern: 1², 2², 3², 4², 5², 6² → Answer: 36

Type 4 — Cube Series:

1, 8, 27, 64, ? → Pattern: 1³, 2³, 3³, 4³, 5³ → Answer: 125

Type 5 — Fibonacci-like Series:

1, 1, 2, 3, 5, 8, ? → Each = sum of previous two → Answer: 13

Type 6 — Alternating Series:

2, 5, 3, 7, 4, 9, ? → Two interleaved series: (2,3,4,?) and (5,7,9,?) → Answer: 5

Type 7 — Increasing Difference Series:

1, 2, 5, 10, 17, ? → Differences: 1, 3, 5, 7, 9 → Answer: 26

Type 8 — Mixed Operation Series:

3, 5, 9, 15, 23, ? → Differences: 2, 4, 6, 8, 10 → Answer: 33

Type 9 — Odd One Out:

2, 3, 5, 7, 11, 14, 17 → All are primes EXCEPT 14 (14 = 2×7, not prime)

Type 10 — Two-Level Series:

4, 9, 20, 43, 90, ?

Differences: 5, 11, 23, 47 → Second-level differences: 6, 12, 24 (×2 pattern) → Next difference = 47×2+1 = 95 → Answer: 90 + 95 = 185

Shortcut: Always compute differences FIRST. If the first-level differences are constant → arithmetic series. If first-level differences are increasing → compute second-level differences. If second-level differences are constant → quadratic series. This 3-step check solves 80% of number series.

7. Ranking Test — Position, Order & Total Count

Ranking tests give you information about positions in a line or queue and ask you to find total count, position from the other end, or positions between people.

📊 Key Ranking Formulas

Formula 1: Total People

Total = Position from Left + Position from Right − 1

Example: Ravi is 7th from the left and 12th from the right. Total = 7 + 12 − 1 = 18 people

Formula 2: Position from Opposite End

Position from Right = Total − Position from Left + 1

Example: In a queue of 30, if Anita is 11th from left, she is 30 − 11 + 1 = 20th from right

Formula 3: People Between Two Persons

If A is at position m from left and B is at position n from left (m < n):

People between A and B = n − m − 1

Example: A is 5th from left, B is 14th from left. Between them = 14 − 5 − 1 = 8 people

Formula 4: When Two People Interchange

If A and B exchange positions, and A's new rank is given, B's new rank = A's old rank.

Ranking Example 1

Problem: In a row of 45 students, Suresh is 20th from the left end. What is his position from the right end?

Solution: Position from right = Total − Position from left + 1 = 45 − 20 + 1 = 26th from right

Ranking Example 2

Problem: Meena is 15th from the left and 18th from the right in a row. How many students are in the row?

Solution: Total = 15 + 18 − 1 = 32 students

Ranking Example 3

Problem: In a queue, A is 10th from the front and B is 8th from the end. If there are 3 people between A and B, how many people are in the queue?

Solution: B's position from front = Total − 8 + 1 = Total − 7. People between A and B = 3. So |B_position − A_position| − 1 = 3 → |B_position − 10| = 4.

Case 1: B_position = 14 → Total − 7 = 14 → Total = 21

Case 2: B_position = 6 → Total − 7 = 6 → Total = 13

Both are valid answers unless additional info is given.

PART III — GUESSTIMATION

8. Market Sizing — Top-Down vs Bottom-Up

Guesstimation is the art of making structured, reasonable estimates for questions that seem impossible to answer precisely. The key word is structured — you need a framework, not a wild guess.

📊 Two Core Approaches

🔽 Top-Down Approach

Start with the BIGGEST number (e.g., total population) and narrow down through filters.

Think of it as: Starting from India's 1.4 billion people and slicing the pie smaller and smaller.

🔼 Bottom-Up Approach

Start with the SMALLEST unit (e.g., one shop, one city block) and scale up.

Think of it as: Counting ATMs on your street, then multiplying by the number of streets, areas, cities.

🔽 Top-Down Example: "How many smartphones are there in India?"

Step 1: India's population = ~1.4 billion = 140 crore

Step 2: Smartphone penetration rate (2024) ≈ 55-60% (not everyone has a phone; children, elderly, rural poor excluded)

Step 3: 140 crore × 0.57 ≈ ~80 crore smartphones

Cross-check: TRAI data says ~85 crore smartphone users in India (2024). Our estimate is within 10% — excellent!

Key assumptions stated: Excluded children below 10 (~15%), elderly without smartphones (~10%), extreme rural (~15%). Some people have 2 phones (dual SIM → ~5% uplift). Net: ~57% penetration.

🔼 Bottom-Up Example: "How many smartphones are there in India?"

Step 1: Average Indian household = 4.5 members

Step 2: Smartphones per household: Urban = ~3, Rural = ~1.5, Average ≈ 2.2

Step 3: Total households in India ≈ 30 crore

Step 4: 30 crore × 2.2 = ~66 crore smartphones

Add dual-SIM/office phones: +15% → ~76 crore

Both approaches give answers in the 70-85 crore range. That's a good sign — the estimates converge!

9. Demand Estimation — Real Indian Examples

Guesstimation Example: "How many Swiggy orders per day in Bangalore?"

Framework: Population → Target Users → Usage Rate → Frequency

Step 1: Bangalore population = ~1.3 crore

Step 2: Who orders on Swiggy?

Working professionals (25-45 yrs): ~30% of population = 39 lakh
College students (18-24 yrs): ~8% = 10.4 lakh
Total target audience: ~50 lakh

Step 3: Smartphone + internet access: ~85% of target = 42.5 lakh

Step 4: Swiggy users (vs Zomato, direct ordering): ~40% market share = 17 lakh active Swiggy users

Step 5: Ordering frequency:

Heavy users (20%): 5 orders/week = ~1.4/day
Medium users (30%): 2 orders/week = ~0.3/day
Light users (50%): 1 order/week = ~0.14/day

Step 6: Calculate daily orders:

= (17L × 20% × 1.4) + (17L × 30% × 0.3) + (17L × 50% × 0.14)

= 4.76L + 1.53L + 1.19L = ~7.5 lakh orders/day

Cross-check: Swiggy processes ~2.5 million orders/day nationwide. Bangalore is their largest city (~25-30%). 25-30% of 25 lakh = 6-7.5 lakh. ✅ Our estimate matches!

10. Revenue/Profit Estimation

Guesstimation Example: "Estimate Zomato's daily revenue"

Step 1: Daily orders

Zomato processes ~2 million orders/day (publicly reported ~Q3 FY24)

Step 2: Average order value (AOV)

Average food order in India: ₹350-450. Let's use ₹400.

Step 3: Revenue streams

Commission from restaurants: ~20-25% of order value = ₹80-100/order
Delivery fee from customer: ~₹30-50/order (avg ₹35)
Platform fee: ₹5-10/order (avg ₹7)
Advertising revenue from restaurants: ~₹5-8/order equivalent

Step 4: Per-order revenue = ₹90 + ₹35 + ₹7 + ₹6 = ~₹138/order

Step 5: Daily revenue = 20 lakh orders × ₹138 = ~₹27.6 crore/day

Step 6: Annualised = ₹27.6 cr × 365 = ~₹10,074 crore/year

Cross-check: Zomato's FY24 revenue was ~₹12,114 crore. Our estimate (₹10,074 cr) is within ~17% — highly respectable for a guesstimation!

11. Product Usage Estimation

Guesstimation Example: "How many Uber rides per day in Delhi?"

Step 1: Delhi NCR population = ~3 crore

Step 2: Target users (use ride-hailing, have smartphones, income > ₹25K/month)

~25% of population = 75 lakh potential users

Step 3: Ride-hailing adoption rate = ~30% of target = 22.5 lakh

Step 4: Uber market share (vs Ola, Rapido) = ~35% = ~8 lakh Uber users

Step 5: Average ride frequency = 2-3 rides/week for active users ≈ 0.36 rides/day

Step 6: Daily rides = 8 lakh × 0.36 = ~2.9 lakh rides/day in Delhi NCR

Sanity check: Uber completes ~10 lakh rides/day across India. Delhi NCR is ~25-30% of India's ride-hailing market. 25-30% of 10L = 2.5-3L rides. ✅ Our answer is spot on!

12. The Guesstimation Master Framework

🏗️ The 5-Step Guesstimation Framework

For ANY guesstimation question, follow this pipeline:

Step 1 → POPULATION BASE

Start with the relevant population. India = 140 crore. A city = X crore. A college = Y students.

Step 2 → SEGMENTATION

Filter the population into relevant segments. Use age, income, geography, behaviour.

Filters: Urban/Rural | Age bracket | Income level | Smartphone access | Interest/need

Step 3 → USAGE RATE

What % of the segment actually uses/buys/does the thing? Not everyone who CAN order food online DOES.

Step 4 → FREQUENCY

How often? Per day, per week, per month, per year? Convert everything to the same time unit.

Step 5 → FINAL NUMBER

Multiply: Base × Segment% × Usage% × Frequency = Answer

Always state your assumptions. Always do a sanity check against known data.

In a consulting interview, the PROCESS matters 10× more than the exact number. An interviewer who asks "How many ATMs in India?" doesn't expect you to know the answer (2.15 lakh). They want to see: (1) Did you break the problem down? (2) Are your assumptions reasonable? (3) Did you do a sanity check? (4) Can you handle pushback? Get the process right, and being off by 30-40% is perfectly fine.

Practice with these Indian guesstimation questions: How many Jio SIM cards are active? How many chai stalls are there in Mumbai? How many IRCTC tickets are booked per day? How many engineers graduate in India every year? Each uses the same 5-step framework.

Section D

3-Tier Practice Labs

🟢 Tier 1 — GUIDED: Syllogism with Venn Diagrams

⏱️ 45–60 minutesBeginnerStep-by-step guided

Task: Solve these 5 syllogism sets by drawing Venn diagrams on paper

Set 1: All doctors are educated. All educated people are literate.

Draw: Doctors ⊂ Educated ⊂ Literate. Conclusions: (a) All doctors are literate ✅ (b) Some literate are doctors ✅ (c) All literate are doctors ❌

Set 2: Some apples are red. All red things are colourful.

Draw: Some Apples overlap Red, Red ⊂ Colourful. Conclusion: Some apples are colourful ✅

Set 3: No fish can fly. All sparrows can fly.

Draw: Fish and Fly are separate. Sparrows ⊂ Fly. Conclusion: No fish are sparrows ✅

Set 4: All roses are flowers. Some flowers are thorny.

Draw: Roses ⊂ Flowers, some Flowers overlap Thorny. Conclusion: "Some roses are thorny" → ❌ No definite conclusion (All + Some = NC)

Set 5: Some Indians are cricketers. Some cricketers are millionaires.

Draw: Indians overlap Cricketers, Cricketers overlap Millionaires. Conclusion: ❌ No conclusion about Indians and Millionaires (Some + Some = NC)

Self-check: Did you get all 5 right? If you missed the traps in Sets 4 and 5, re-read Section C.4 (Combination Rules).

🟡 Tier 2 — SEMI-GUIDED: Guesstimation Practice

⏱️ 60–90 minutesIntermediateFramework given, you fill the numbers

Task: Use the 5-step framework to answer these 3 questions

Question 1: How many barbers/salons are there in Pune?

Hints: Pune population ≈ 70 lakh. Average haircut frequency for men = once/month, women = once/2 months. A barber serves ~15 customers/day. Work 26 days/month.

Question 2: How many litres of milk are consumed in India per day?

Hints: India population = 140 crore. Average milk consumption per capita per day ≈ 400-450 ml (India is the world's largest milk producer). Not everyone drinks milk — vegetarians more, some are lactose intolerant.

Question 3: Estimate the number of auto-rickshaws in Bangalore.

Hints: Think bottom-up. How many autos do you see per km of road? Bangalore has ~14,000 km of road. But autos only run on ~30% of roads (not highways, residential lanes).

Stretch goal: Compare your answers with real data (Google it after your estimation). If you're within 2× of the actual number, you're doing great. Within 50% = consulting-ready!

🔴 Tier 3 — OPEN CHALLENGE: Create Your Own Guesstimation Portfolio

⏱️ 2–3 hoursAdvancedNo instructions — real consulting prep

The Brief:

Create a "Guesstimation Portfolio" document with 5 fully worked-out guesstimation problems relevant to Indian markets. For each:

State the question clearly
Choose top-down or bottom-up (or both) approach
Show all 5 framework steps with calculations
State every assumption explicitly
Do a sanity check with real data
Provide a confidence range (e.g., "I estimate 50,000–80,000 with most likely value 65,000")

Suggested questions:

How many wedding photographers are there in India?
How many Paytm transactions happen per day?
How many pizzas does Domino's sell per day in India?
How many engineering colleges are there in India?
How many Ola/Uber drivers are there in Hyderabad?

This portfolio is a REAL asset. Consulting firms look for "structured thinking" evidence. A well-done guesstimation portfolio shared on LinkedIn or during interviews immediately signals analytical ability. Several students have landed McKinsey/BCG interview calls based on guesstimation content they shared.

Section E

Problem Set — Mixed Practice

Syllogism Problems

P1. Statements: All pens are pencils. Some pencils are erasers.

Conclusions: I. Some pens are erasers. II. Some erasers are pens.

Answer: Neither I nor II follows (All + Some = No Conclusion)

P2. Statements: No car is a bus. All buses are vehicles.

Conclusions: I. No car is a vehicle. II. Some vehicles are not cars.

Answer: Only II follows. (Cars COULD be vehicles through another path; but some vehicles = buses ≠ cars)

P3. Statements: All laptops are gadgets. All phones are gadgets. Some gadgets are expensive.

Conclusions: I. Some laptops are phones. II. Some gadgets are laptops.

Answer: Only II follows. (All laptops are gadgets → Some gadgets are laptops ✅. Laptops and phones are both gadgets but may not overlap ❌)

Number Series Problems

P4. 2, 6, 12, 20, 30, ? → Differences: 4, 6, 8, 10, 12 → Answer: 42

P5. 1, 4, 27, 256, ? → Pattern: 1¹, 2², 3³, 4⁴, 5⁵ → Answer: 3125

P6. 3, 5, 9, 17, 33, ? → Differences: 2, 4, 8, 16, 32 → Answer: 65

Ranking Problems

P7. In a class of 40, Rohit is 15th from the top. What is his rank from the bottom?

Answer: 40 − 15 + 1 = 26th from bottom

P8. Priya is 9th from the left and 15th from the right. How many people are in the row?

Answer: 9 + 15 − 1 = 23 people

Guesstimation Problems

P9. Estimate the number of petrol pumps in India. (Hint: Think about districts, highways, city density)

Approach: India has ~780 districts. Average ~80-100 petrol pumps per district. Total ≈ 65,000-80,000. Actual: ~84,000 (2024). ✅

P10. How many litres of Coca-Cola are sold in India per day?

Approach: India population 140 cr. Carbonated drink penetration ~15-20%. Average consumption: ~100ml per consumer per week. Daily = 140cr × 17% × 100ml ÷ 7 ≈ 34 crore ml = 3.4 crore litres/day for all brands. Coca-Cola market share ~60% → ~2 crore litres/day.

Section F

MCQ Assessment Bank — 30 Questions (Bloom's Mapped)

Remember / Recall (Q1–Q5)

The statement "All A are B" is classified as which type of proposition?

Particular Affirmative
Universal Affirmative
Universal Negative
Particular Negative

Remember

✅ Answer: (B) Universal Affirmative — "All A are B" is a Type A (Universal Affirmative) proposition where every member of A belongs to B.

In a Venn diagram, "No A are B" is represented by:

A inside B
A and B overlapping
A and B completely separate
A and B identical

Remember

✅ Answer: (C) — "No A are B" means the two sets have zero intersection, shown as two separate non-touching circles.

What is the formula for total number of people when a person is Rth from the right and Lth from the left?

R + L
R + L − 1
R + L + 1
R × L

Remember

✅ Answer: (B) R + L − 1 — The person is counted once in both positions, so we subtract 1 to avoid double-counting.

In guesstimation, the "top-down" approach starts with:

The smallest unit and scales up
The total population/market and narrows down
Revenue data from company reports
Random assumptions

Remember

✅ Answer: (B) — Top-down starts with the biggest number (e.g., India's total population) and applies successive filters to narrow down to the answer.

Which combination of premises gives NO valid conclusion in syllogism?

All + All
All + No
Some + Some
Some + All

Remember

✅ Answer: (C) Some + Some — When both premises are particular (Some), no definite conclusion can be drawn about the relationship between the two end terms.

Understand / Explain (Q6–Q10)

Why does "All A are B" NOT imply "All B are A"?

Because logic is always reversible
Because A is a subset of B, but B may contain elements outside A
Because A and B are always equal sets
Because the statement is always false

Understand

✅ Answer: (B) — "All A are B" means A ⊂ B. B can have members that are NOT in A. Example: All dogs are animals, but not all animals are dogs.

In a ranking problem, if the total number of students is N and a student is Kth from the top, which expression gives the rank from the bottom?

N − K
N − K + 1
N + K − 1
K − N + 1

Understand

✅ Answer: (B) N − K + 1 — If a student is Kth from top in a class of N, the position from bottom = N − K + 1.

Why is stating assumptions important in guesstimation?

It makes the answer look longer
It shows the interviewer your thought process and allows them to challenge specific assumptions
It's not important — only the final number matters
It helps you memorise formulas

Understand

✅ Answer: (B) — In consulting interviews, stating assumptions transparently shows structured thinking and gives the interviewer specific points to probe. The process matters more than the number.

What does a "complementary pair" mean in syllogism?

Two conclusions that are both true
Two conclusions that are both false
Two conclusions where exactly one MUST be true
Two conclusions that contradict the premises

Understand

✅ Answer: (C) — A complementary pair consists of two conclusions (like "Some A are B" and "Some A are not B") where at least one must necessarily be true, even if we can't determine which one.

Q10

In the number series 2, 6, 18, 54, ?, the pattern is:

Adding 4 each time
Multiplying by 3 each time
Squaring each term
Adding consecutive primes

Understand

✅ Answer: (B) — This is a geometric series with common ratio 3. Each term = previous × 3. Next term = 54 × 3 = 162.

Apply / Solve (Q11–Q15)

Q11

Statements: All bags are boxes. All boxes are cartons. Conclusion: All bags are cartons.

Follows
Does not follow
Can't determine
Data insufficient

Apply

✅ Answer: (A) Follows — All + All = All. Bags ⊂ Boxes ⊂ Cartons → Bags ⊂ Cartons.

Q12

In a row of 50 students, Ram is 14th from the left. How many students are to his right?

Apply

✅ Answer: (B) 36 — Students to his right = Total − Position from left = 50 − 14 = 36.

Q13

Find the next number: 5, 11, 23, 47, ?

Apply

✅ Answer: (C) 95 — Pattern: each term = previous × 2 + 1. So 47 × 2 + 1 = 95.

Q14

Statements: Some cats are dogs. No dogs are rats. Conclusion: Some cats are not rats.

Follows
Does not follow
Either follows or doesn't
Data insufficient

Apply

✅ Answer: (A) Follows — Some + No = Some Not. The cats that ARE dogs cannot be rats (since no dogs are rats). Hence some cats are not rats.

Q15

Using top-down approach: India has 140 crore people, 30% are in the 20-40 age group, 60% of them have smartphones, and 25% use dating apps. How many dating app users?

4.2 crore
6.3 crore
8.4 crore
3.15 crore

Apply

✅ Answer: (B) 6.3 crore — 140 × 0.30 × 0.60 × 0.25 = 6.3 crore dating app users.

Analyse / Compare (Q16–Q20)

Q16

Statements: All roses are flowers. Some flowers are red. Can we conclude "Some roses are red"?

Definitely true
Definitely false
Possibly true but not definite
The premises are contradictory

Analyse

✅ Answer: (C) — All + Some = No definite conclusion. The "some flowers that are red" might be flowers OTHER than roses. It CAN be true (possibility) but is NOT necessarily true.

Q17

Which number doesn't belong: 1, 4, 9, 16, 20, 25, 36?

Analyse

✅ Answer: (C) 20 — All others are perfect squares (1², 2², 3², 4², 5², 6²). 20 is not a perfect square.

Q18

A guesstimation gives "5 crore users" for a niche B2B SaaS product in India. What's the most likely error?

Using bottom-up instead of top-down
Overestimating the target market — 5 crore is too high for B2B niche
Underestimating smartphone penetration
Using too many decimal places

Analyse

✅ Answer: (B) — A "niche B2B SaaS" product would likely have thousands to a few lakhs of users, not crores. 5 crore suggests the analyst didn't properly segment the B2B market from the general consumer market.

Q19

Compare: Arithmetic series 2, 5, 8, 11... vs Geometric series 2, 6, 18, 54... Which grows faster?

Arithmetic always grows faster
Geometric always grows faster after initial terms
They grow at the same rate
It depends on the starting number

Analyse

✅ Answer: (B) — Geometric series grows exponentially while arithmetic grows linearly. After initial terms, geometric always dominates (e.g., by term 10: arithmetic = 29 vs geometric = 39,366).

Q20

A student solves a ranking problem: "X is 8th from left, Y is 12th from right, total = 8 + 12 = 20." What's wrong?

Nothing is wrong
The formula should be 8 × 12
The formula should be 8 + 12 − 1 = 19, since the person at position is counted in both
The positions should be averaged

Analyse

✅ Answer: (C) — The correct formula is Total = L + R − 1. The student forgot to subtract 1 (the person themselves is counted from both ends). If X = Y, total = 8 + 12 − 1 = 19. If X ≠ Y, we need more information.

Evaluate / Assess (Q21–Q25)

Q21

A student claims: "Since All A are B and All A are C, therefore All B are C." Is this valid?

Yes, it's a valid logical deduction
No — A being inside both B and C doesn't mean B and C overlap
Yes, by the transitive property
It depends on the number of elements in A

Evaluate

✅ Answer: (B) — This is INVALID. A can be a subset of both B and C without B and C having any relationship. Example: All roses are flowers AND all roses are red. But NOT all flowers are red.

Q22

A guesstimation for "number of restaurants in India" arrives at 50 lakh. The actual number is ~75 lakh. How would you rate this estimate?

Very poor — more than 10% off
Poor — more than 50% off
Good — within 2× of actual, acceptable for guesstimation
Perfect — no improvement needed

Evaluate

✅ Answer: (C) — In guesstimation, being within 2× (half to double) of the actual answer is considered acceptable. 50L vs 75L is within 1.5× — this is a GOOD estimate. The key is the process, not exact precision.

Q23

Evaluate this syllogism solution: "All cats are animals. Some animals are pets. ∴ Some cats are pets." Is the student correct?

Yes, the conclusion is valid
No — "All + Some" gives no definite conclusion; the "some animals" that are pets might not be cats
No — the conclusion should be "All cats are pets"
Yes — because cats are commonly pets

Evaluate

✅ Answer: (B) — This is the classic All + Some = No Conclusion trap. The "some animals that are pets" could be dogs, birds, etc. — not necessarily cats. Real-world knowledge (cats ARE often pets) must NOT influence logical deduction.

Q24

Which odd-one-out answer is more defensible: "Remove 15 from {3, 5, 7, 11, 13, 15}" or "Remove 3 from the same set"?

Remove 3 — it's the smallest
Remove 15 — it's the only non-prime; all others are prime numbers
Both are equally valid
Neither — the set has no odd one out

Evaluate

✅ Answer: (B) — 3, 5, 7, 11, 13 are all prime numbers. 15 = 3 × 5 is composite (not prime). The mathematical property "prime vs non-prime" is stronger than "smallest number."

Q25

A candidate estimates "10 lakh ATMs in India." The actual number is ~2.15 lakh. What went wrong?

They probably confused bank branches with ATMs
Their population segmentation was correct
They used bottom-up approach perfectly
Nothing — this is within acceptable range

Evaluate

✅ Answer: (A) — 10 lakh is ~5× the actual number, which is outside the acceptable 2× range. The candidate likely confused ATMs with bank branches (~1.5 lakh branches exist) or didn't apply proper filters for ATM-per-population ratio.

Create / Design (Q26–Q30)

Q26

Design a syllogism where "All + All" leads to a conclusion but the converse does NOT hold. Which set of statements achieves this?

All X are Y. All Y are Z. ∴ All X are Z (but NOT All Z are X)
Some X are Y. Some Y are Z. ∴ Some X are Z
No X are Y. All Y are Z. ∴ No X are Z
All X are X. ∴ All X are X

Create

✅ Answer: (A) — "All X are Y, All Y are Z → All X are Z" is valid. But "All Z are X" is NOT valid because Z may have elements outside Y, and Y may have elements outside X.

Q27

Create a number series where the pattern involves alternating +3 and ×2: Starting with 1, what is the 6th term?

Create

✅ Answer: (C) 22 — Series: 1 (+3)→ 4 (×2)→ 8 (+3)→ 11 (×2)→ 22 (+3)→ 25. Wait, the 6th term: 1, 4, 8, 11, 22, 25. The 5th term is 22 and the 6th is 25. Re-counting: T1=1, T2=4, T3=8, T4=11, T5=22, T6=25. Answer is (A) 25 for the 6th term. But if "starting with 1" means T1=1 and we want T6, it's 25. Accept (A).

Q28

Which framework step would you add FIRST to improve a guesstimation for "number of gyms in India"?

Calculate gym equipment cost
Segment by tier-1 vs tier-2 vs tier-3 cities, since gym density varies dramatically
Look up stock prices of gym chains
Count the number of fitness influencers on Instagram

Create

✅ Answer: (B) — City-tier segmentation is crucial because tier-1 cities (Mumbai, Delhi) might have 1 gym per 2,000 people, while rural areas might have 1 per 50,000. Without this segmentation, the estimate will be wildly off.

Q29

Create a 3-premise syllogism that leads to the conclusion "Some doctors are not lazy." Which set works?

All doctors are hardworking. No hardworking person is lazy. (Only 2 premises needed!)
Some doctors are tall. Some tall people are lazy. (Some + Some = NC)
All lazy people are doctors. No doctors are hardworking. (Leads to different conclusion)
Some doctors are women. All women are lazy. (Contradicts the target conclusion)

Create

✅ Answer: (A) — "All doctors are hardworking" + "No hardworking person is lazy" → "No doctors are lazy" (which is even stronger than "Some doctors are not lazy"). This works with just 2 premises!

Q30

Design a bottom-up estimation framework for "How many chai stalls in Mumbai?" Which starting unit is best?

Start with India's GDP
Start with one residential locality — count chai stalls per 500m street, then scale by total streets/areas in Mumbai
Start with annual tea consumption in kg
Start with Starbucks store count

Create

✅ Answer: (B) — Bottom-up means starting from the smallest observable unit. Counting chai stalls per 500m of street in a typical area, then multiplying by total commercial streets in Mumbai gives the most reliable ground-up estimate.

Section G

Short Answer Questions (8 Questions)

SA-1: Explain the 4 standard propositions of syllogism with one example each. (4 marks)

Model Answer:

1. Universal Affirmative (All A are B): Every member of A belongs to B. Example: "All engineers study mathematics." Venn: A circle inside B circle.

2. Universal Negative (No A are B): No member of A belongs to B. Example: "No vegetarian eats chicken." Venn: Two separate circles.

3. Particular Affirmative (Some A are B): At least one member of A belongs to B. Example: "Some politicians are honest." Venn: Overlapping circles.

4. Particular Negative (Some A are not B): At least one member of A does not belong to B. Example: "Some students are not regular." Venn: Part of A outside B.

SA-2: What is the difference between "must be true" and "can be true" in syllogism? (3 marks)

Model Answer:

"Must be true" means the conclusion follows necessarily from the premises in EVERY possible valid Venn diagram. There is no way to draw the diagram where the conclusion fails. Example: "All A are B, All B are C → All A are C" MUST be true.

"Can be true" means the conclusion is NOT contradicted by the premises — there exists at least ONE valid diagram where it holds, even if there are other diagrams where it doesn't. Example: "All A are B → All B are A" CAN be true (if A = B) but need not be.

SA-3: State the formula for finding total people in a linear arrangement. Solve: A is 13th from left, 20th from right. (2 marks)

Model Answer:

Formula: Total = Position from Left + Position from Right − 1

Total = 13 + 20 − 1 = 32 people

SA-4: Explain top-down vs bottom-up approach in guesstimation with a simple example. (4 marks)

Model Answer:

Top-Down: Start with the largest number and filter down. "How many restaurants in Jaipur?" → Start: Jaipur population (40 lakh) → 1 restaurant per ~200 people → ~20,000 restaurants.

Bottom-Up: Start with the smallest unit and scale up. Count restaurants on one main road (~15 per km), estimate total commercial road length in Jaipur (~1,500 km), but only 30% have restaurants → 15 × 450 ≈ 6,750. Add residential area eateries → ~15,000-20,000.

When both approaches give similar numbers (15,000-20,000), you can be confident in your estimate.

SA-5: Find the odd one out and explain: 2, 3, 5, 7, 9, 11, 13 (2 marks)

Model Answer:

The odd one out is 9. All others (2, 3, 5, 7, 11, 13) are prime numbers. 9 = 3 × 3 is a composite number, not prime.

SA-6: Why does "Some + Some = No Conclusion" in syllogism? Illustrate with a Venn diagram description. (3 marks)

Model Answer:

When both premises are particular ("Some A are B, Some B are C"), the overlapping parts of A with B, and B with C, may not share any common elements. The "some A" that are B might be completely different from the "some B" that are C.

Imagine three circles: A, B, C. A overlaps with the LEFT side of B. C overlaps with the RIGHT side of B. A and C never touch! Hence, we CANNOT conclude anything about the relationship between A and C. The two "some" overlaps are independent.

SA-7: In a guesstimation, what is a "sanity check" and why is it important? (3 marks)

Model Answer:

A sanity check is a quick cross-verification of your guesstimation answer against known data, common sense, or an alternative approach.

Methods: (1) Compare with publicly reported data (e.g., company annual reports). (2) Use an alternative approach (if you did top-down, try bottom-up). (3) Per-capita check — does the per-person number make sense?

Why important: It catches order-of-magnitude errors (e.g., estimating 50 crore ATMs instead of 2 lakh). In consulting interviews, interviewers expect candidates to self-check their answers.

SA-8: Identify the next 2 terms: 1, 1, 2, 3, 5, 8, 13, ?, ? and name the series. (2 marks)

Model Answer:

This is the Fibonacci series. Each term = sum of the two preceding terms.

13 + 8 = 21

21 + 13 = 34

Next two terms: 21 and 34.

Section H

Long Answer / Case Study Questions (3 Questions)

LA-1: Complete Syllogism Analysis (10 marks)

Question: Given the following premises, draw Venn diagrams and evaluate ALL conclusions:

Premises:

All MBA graduates are managers.
Some managers are leaders.
No leaders are followers.

Evaluate these conclusions:

I. Some MBA graduates are leaders.

II. No MBA graduates are followers.

III. Some managers are not followers.

IV. All MBA graduates are followers. (Possibility check)

Model Answer:

Venn Diagram: MBA ⊂ Managers. Managers partially overlaps Leaders. Leaders is completely separate from Followers.

I. "Some MBA graduates are leaders" → ❌ Does NOT follow. All + Some = No Conclusion. The "some managers who are leaders" may be managers who are NOT MBA graduates. The MBA circle sits inside Managers, but the Leaders overlap might be on the other side of Managers.

II. "No MBA graduates are followers" → ❌ Cannot be definitively concluded. We can trace MBA → Managers → (some) Leaders → NOT Followers. But MBA graduates who are NOT leaders could still potentially be followers. However, we need to check: is there a path from MBA to Followers? Since All + Some = NC for MBA-Leaders, and Leaders are separate from Followers, we can't establish a definite link either way.

III. "Some managers are not followers" → ✅ FOLLOWS. Some managers ARE leaders (Premise 2). No leaders are followers (Premise 3). Therefore, those managers who are leaders are definitely NOT followers. Hence, some managers are not followers.

IV. "All MBA graduates are followers" — Possibility check → ✅ CAN be true as a possibility. Since we can't definitively link MBA to Leaders (All+Some=NC), MBA graduates might all fall in the non-leader part of Managers. And since the relationship between non-leader Managers and Followers is unspecified, MBA graduates COULD potentially be followers. However, this would only work if MBA graduates and Leaders don't overlap within Managers.

LA-2: Full Guesstimation — "How many ATMs are there in India?" (10 marks)

Question: Using the 5-step guesstimation framework, estimate the total number of ATMs in India. Show your work with the top-down approach, state all assumptions, perform a sanity check, and provide a confidence range.

Model Answer:

Step 1 — Population Base: India population = 140 crore. But ATMs serve bank account holders, not total population. Number of bank accounts ≈ 150 crore (many have multiple accounts due to Jan Dhan Yojana).

Step 2 — Segmentation:

Urban India: ~35% of population = 49 crore people, higher ATM density
Semi-urban: ~20% = 28 crore people
Rural: ~45% = 63 crore people, very low ATM density

Step 3 — ATM Density Estimation:

Urban: ~1 ATM per 3,000 people → 49,00,00,000 ÷ 3,000 = ~1,63,000 ATMs
Semi-urban: ~1 ATM per 8,000 people → 28,00,00,000 ÷ 8,000 = ~35,000 ATMs
Rural: ~1 ATM per 25,000 people → 63,00,00,000 ÷ 25,000 = ~25,200 ATMs

Step 4 — Total Estimate: 1,63,000 + 35,000 + 25,200 = ~2,23,200 ATMs

Step 5 — Sanity Check:

RBI data (March 2024): ~2,15,000 ATMs in India
Our estimate: 2,23,200 — within 4% of actual! ✅ Excellent
Alternative check: India has ~12 ATMs per 1 lakh population (World Bank data). 140 crore × 12/1,00,000 = 1,68,000. Slightly lower because World Bank uses total population, not just banked.

Confidence Range: 1,80,000 – 2,50,000 with most likely value ~2,20,000

Key Assumptions: (1) Urban-rural population split of 35-20-45. (2) ATM density decreases dramatically from urban to rural. (3) Multiple ATMs at railway stations, malls, airports counted in urban. (4) White-label ATMs (Tata, Hitachi) included.

LA-3: Integrated Problem — Syllogism + Ranking + Estimation (10 marks)

Question: A company has 60 employees standing in a queue for a team-building exercise.

Part A (Ranking — 3 marks): Manager Arjun is 22nd from the front. Manager Kavita is 35th from the back. How many people are between them? Is Arjun in front of Kavita or behind?

Part B (Syllogism — 4 marks): The company has these rules:
1. All managers are MBA graduates.
2. Some MBA graduates are IIT alumni.
3. No IIT alumni work in sales.
Can we conclude: "Some managers do not work in sales"? Draw Venn diagram and prove.

Part C (Guesstimation — 3 marks): The company is a fintech startup with 60 employees. Estimate their monthly AWS cloud bill.

Model Answer:

Part A:

Kavita's position from front = Total − Position from back + 1 = 60 − 35 + 1 = 26th from front.

Arjun = 22nd, Kavita = 26th. Arjun is in front of Kavita.

People between = 26 − 22 − 1 = 3 people between them.

Part B:

Managers ⊂ MBA (Premise 1). MBA partially overlaps IIT (Premise 2). IIT is separate from Sales (Premise 3).

From Premises 2 + 3: Some MBA are IIT → No IIT are Sales → Some MBA are not in Sales ✅

But from Premises 1 + (2+3): All Managers are MBA + Some MBA are not Sales → All + Some = No definite conclusion about Managers and Sales.

So we CANNOT definitively conclude "Some managers do not work in sales." ❌

However, it "CAN be true" as a possibility. ✅

Part C:

A 60-person fintech startup likely has: ~10-15 microservices, database clusters, CI/CD pipeline, monitoring.

EC2/ECS compute: ~$3,000-5,000/month (15-20 servers)
RDS databases: ~$1,000-2,000/month
S3 storage: ~$200-500/month
Other (Lambda, CloudWatch, SES): ~$500-1,000/month

Estimated monthly AWS bill: $5,000-$8,500/month (₹4-7 lakh/month)

Sanity check: Industry benchmark for 50-100 person fintech = $3,000-$15,000/month. Our estimate falls in the middle. ✅

Section I

Industry Spotlight — A Day in the Life

👩‍💼 Neha Kapoor, 28 — Management Consultant at Bain & Company, Mumbai

Background: B.Com from SRCC, Delhi University. MBA from IIM Lucknow. No engineering background. Self-studied analytical reasoning during CAT prep. Joined Bain as an Associate Consultant after campus placement.

A Typical Day:

8:30 AM — Morning check-in with the case team. Currently working on a market entry strategy for a European FMCG brand entering India.

9:00 AM — "How large is the premium shampoo market in India?" — Neha builds a guesstimation model. Population → urban women 20-50 years → income > ₹50K/month → willing to pay ₹500+ for shampoo → usage frequency → market size in ₹ crores.

11:00 AM — Analyse competitor data. Use logical frameworks to categorise competitors: domestic (Marico, Dabur) vs. international (P&G, Unilever). Syllogism-like thinking: "All premium brands need urban distribution. This brand lacks urban distribution. Therefore..."

1:00 PM — Client lunch at BKC (Bandra Kurla Complex). Present initial market sizing findings. Client pushes back on rural assumptions — Neha adjusts the model live.

3:00 PM — Build Excel models. Revenue projections for 5 years across 15 product SKUs.

5:00 PM — Interview prep — Neha is now an interviewer! She designs guesstimation questions for incoming candidates: "How many washing machines are sold in India per year?"

7:00 PM — Team dinner. Case wrap-up in 2 weeks. The client will pay ₹3-5 crore for this 8-week engagement.

Detail	Info
Skills Used Daily	Guesstimation, logical reasoning, Excel modelling, structured communication, data analysis
Entry Salary (AC role)	₹22-28 LPA + bonus
Post-MBA (3-5 yrs)	₹35-50 LPA
Partner Level (15+ yrs)	₹2-5 crore/year
Firms Hiring	McKinsey, BCG, Bain, Kearney, Strategy&, Deloitte Strategy, Accenture Strategy
Key Skill from This Chapter	Guesstimation (used daily) + Logical reasoning (used in case structuring)

You don't need an IIM degree to use these skills. Guesstimation and logical reasoning are tested in KPO (Knowledge Process Outsourcing) firms, analytics companies, and even product management roles at startups. Companies like Mu Sigma, ZS Associates, and Fractal Analytics hire BBA/BCA graduates and test these exact skills.

Section J

Earn With It — Consulting Prep Coaching

💰 Your Earning Path After This Chapter

Portfolio Piece: A "Guesstimation Portfolio" with 5 fully worked-out Indian market-sizing problems + a solved syllogism set of 50 questions.

Earning Opportunities:

• Peer tutoring for Bank PO/SSC Reasoning — ₹200-500/hour per student

• CAT coaching centre assistant (reasoning section) — ₹5,000-10,000/month part-time

• Create a YouTube/Instagram channel on "Guesstimation for Indian Markets" — ad revenue + sponsorships

• Sell guesstimation practice sets on Gumroad/Instamojo — ₹199-499 per set

• Campus placement prep coaching for juniors — ₹2,000-5,000 per student

Earning Avenue	Investment Needed	Monthly Potential
Peer tutoring (reasoning)	₹0 (just your knowledge)	₹3,000–₹10,000
Bank PO batch teaching	₹0 (use college classrooms)	₹5,000–₹15,000
Guesstimation content creation	Phone + free editing tools	₹2,000–₹20,000 (grows over time)
Placement prep packages	₹0	₹5,000–₹25,000 (seasonal)
Practice set sales (digital)	₹0 (use free platforms)	₹1,000–₹5,000 (passive)

⏱️ Time to First Earning: 1–2 weeks (start by helping 3 juniors with Bank PO reasoning prep)

The multiplier effect: Once you've mastered guesstimation, you can create content that earns passively. A single well-made YouTube video on "How to estimate the number of ATMs in India" can get 10,000+ views from CAT/consulting aspirants and earn indefinitely through ads.

Section K

Chapter Summary — Quick Revision

🗂️ Key Takeaways

SYLLOGISM:

4 proposition types: All (A), No (E), Some (I), Some Not (O)
Always draw Venn diagrams — never solve in your head
All + All = All ✅ | All + Some = NC ❌ | Some + Some = NC ❌ | Some + No = Some Not ✅
"Can be true" ≠ "Must be true" — possibility questions need only ONE valid diagram
Watch for complementary pairs: "Some A are B" OR "Some A are not B" — one MUST be true
Never use real-world knowledge to override logical deduction

NUMBER RANKING:

Number series: Check differences first (constant = arithmetic, increasing = check second-level)
10 types: Arithmetic, Geometric, Square, Cube, Fibonacci, Alternating, Mixed, Two-level, Odd-one-out
Ranking: Total = Left + Right − 1 | Position from other end = Total − Position + 1
Between two people = |Pos1 − Pos2| − 1

GUESSTIMATION:

5-step framework: Population → Segmentation → Usage Rate → Frequency → Final Number
Two approaches: Top-down (big to small) and Bottom-up (small to big)
Always state assumptions explicitly
Always do a sanity check
Within 2× of actual = good. Within 50% = excellent. Process > precision.

Remember the Aadhaar analogy: "All Indian citizens should have Aadhaar, but all Aadhaar holders are not citizens." This single sentence captures the essence of syllogism — the direction of "All" matters. Master this concept, and you'll never make the converse error again.

Section L

Earning Checkpoint — Self-Assessment

Skill	Tool / Method	Deliverable	Earning-Ready?
Syllogism (4 propositions)	Pen + Paper / Venn Diagrams	Solved 50+ problems accurately	✅ Yes — can tutor Bank PO reasoning
Possibility-based questions	Decision framework	Can distinguish must/can/false	✅ Yes — advanced coaching material
Number series	Pattern recognition	Can solve all 10 types	✅ Yes — placement prep content
Ranking problems	Formulas + logic	Sub-30-second solve time	✅ Yes — speed drilling for students
Guesstimation (top-down)	5-step framework	5 Indian market-sizing problems	✅ Yes — consulting prep coaching
Guesstimation (bottom-up)	Unit economics → scale	3 bottom-up estimations	✅ Yes — LinkedIn content creation
Sanity checking	Cross-verification	All estimates within 2× of actual	✅ Yes — consulting-interview ready

Minimum Viable Earning Setup after this chapter: A set of 50 solved syllogism problems + 5 guesstimation case studies + ability to teach reasoning = you can earn ₹3,000–₹15,000/month through tutoring, content creation, or digital product sales while still in college.

✅ Unit 2 complete. Ready for Unit 3!

[QR: Link to EduArtha video tutorial — Syllogism, Ranking & Guesstimation]