Module 1 · IBPS PO Reasoning Ability

Series & Analogy

Updated 17 Nov 2026 · 4,099 words · illustrated
Updated 21 Nov 2026
Number series, alphabet series, word and letter analogies.
Number Series · Alphabet Series · Word Analogies · Letter Analogies
🎯 Take Mock Test
Number SeriesPatternsLetter SeriesPatternsFigure SeriesPatternsArithmeticProgression (+/-constant)GeometricProgression (*/÷constant)Prime / Square /Cube SeriesAlternating / MixedOperationsAlphabetical Order(Forward/Backward)Skipping Letters(+1, +2, +3...)Positional Patterns(Vowels,Consonants)Cyclic / RepeatingSequencesRotation /Reflection (90°,180°, mirror)Addition / Deletionof ElementsShapeTransformation(Size, Shading)Positional Shift(Clockwise, Grid)Exam Hook: Identify pattern type first → apply rule → verify with 2+ termsGeography link: River tributary order, mountain range elevation series, climate zone sequencesvlymbooq
Series Analogy Pattern Types for Competitive Exams — Visual comparison of the three main series analogy categories and their common patterns for quick pattern recognition. · Open the full diagram →

Learning Objectives

  • Understand the fundamental concepts of Series & Analogy
  • Apply key formulas and techniques to solve problems
  • Practice with exam-level questions to build speed and accuracy

What are Series and Analogy?

Series and Analogy questions in banking exams test your pattern recognition abilities. In series questions, you identify the next term or missing term in a sequence of letters or numbers based on a hidden rule. In analogy questions, you find the relationship between a pair and apply it to another pair. These questions are quick-scoring and appear in IBPS PO prelims.

Types of Series Questions

1. Letter Series: A sequence of letters is given with a pattern based on positional values (A=1 to Z=26). Patterns include addition/subtraction of fixed values, skipping letters, reversal, or a combination. Example: A, C, F, J, ? → A(+2)→C(+3)→F(+4)→J(+5)→O.

2. Number Series: A sequence of numbers follows a pattern — arithmetic progression, geometric progression, squares, cubes, prime numbers, Fibonacci, or mixed operations. Example: 2, 6, 12, 20, ? → 1×2, 2×3, 3×4, 4×5, 5×6 = 30.

3. Alpha-Numeric Series: Combination of letters and numbers with a pattern in both. Example: A1, C4, E9, G16, ? → Letters: A→C→E→G→I (+2 each). Numbers: 1,4,9,16,25 (squares). Next: I25.

Types of Analogy Questions

1. Word Analogy: Relationship between words — synonym, antonym, cause-effect, worker-tool, animal-young one, etc.

2. Letter Analogy: Based on positional values or alphabetical patterns. Example: ACE : FHJ :: MOQ : ?

3. Number Analogy: Based on arithmetic operations. Example: 4 : 20 :: 6 : ?

Solved Examples

Example 1 (Letter Series)
Find the next term: B, E, H, K, N, ?

a) O b) P c) Q d) R
Solution: c) Q. B=2, E=5, H=8, K=11, N=14. The pattern is +3 each step. Next = 14+3 = 17 → Q.
Example 2 (Number Analogy)
Find the missing term: 3 : 27 :: 5 : ?

a) 75 b) 125 c) 100 d) 150
Solution: b) 125. The relationship is n : n³. 3³ = 27. 5³ = 125.
Example 3 (Alpha-Numeric Series)
Find the next term: D3, G6, J9, M12, ?

a) P15 b) O15 c) P14 d) O14
Solution: a) P15. Letters: D→G→J→M→P (+3 each). Numbers: 3,6,9,12,15 (+3 each). Next: P15.

Shortcut Techniques

  • For letter series, convert letters to numbers (A=1 to Z=26) to see the arithmetic pattern clearly.
  • For number series, check differences first. If differences are constant, it's AP. If differences increase/decrease by a fixed amount, it's a quadratic pattern.
  • For analogies, identify the exact relationship — is it synonym, antonym, part-whole, cause-effect, or mathematical?
  • Practice with mixed patterns — some series combine two patterns alternatingly.
Pro Tip
In IBPS PO, series questions often have complex patterns like '+3, ×2, +3, ×2' or '×1+1, ×2+2, ×3+3'. Always check for alternate operation patterns when the simple ones don't work.

Advanced Problem Types

Competitive exams feature advanced variations combining multiple reasoning concepts. Master these to score high.

Multi-Step Problems

Some questions require applying multiple reasoning techniques sequentially. Break problems into clear steps.

Common Traps

  • Assuming a pattern continues without checking all elements
  • Misinterpreting negative statements in syllogisms
  • Forgetting wrapping in alphabet coding (Z to A transition)
  • Confusing directions (East vs West)
  • Overlooking either-or cases in syllogisms
  • Not considering all possible arrangements in puzzles
Pro Tip
For puzzles, start with the most concrete clue and build around it. Draw diagrams for seating arrangements and family trees for blood relations.

Recursive Difference Patterns in Number Series

Number series in competitive reasoning often hide a recursive rule where each term is generated from one or more preceding terms. The simplest recursion is an arithmetic progression, but examiners frequently employ second‑order differences, multiplicative recursions, or mixed additive‑multiplicative rules. A second‑order difference series shows a constant difference between successive first differences; for example, the series 2, 5, 10, 17, 26 has first differences 3, 5, 7, 9 and a constant second difference of 2, revealing a quadratic formula n²+1. When the second difference itself changes by a fixed amount, the underlying polynomial rises to cubic order. Multiplicative recursions appear as geometric progressions (×r) or as patterns like ×2+1, ×3+2, ×4+3 where the multiplier and the addend both increase linearly. A mixed recursion may alternate between two distinct rules, such as +3, ×2, +3, ×2, which produces 4, 7, 14, 17, 34, 37. Recognising the alternation requires writing the operation sequence explicitly and checking whether the pattern holds for every adjacent pair. Examiners also embed hidden base cases: a series may start with two seed values and then follow a Fibonacci‑type addition (Tₙ = Tₙ₋₁ + Tₙ₋₂) or a Tribonacci rule (sum of three previous terms). In such cases the first few terms look arbitrary until the recursion becomes apparent. The key technique is to compute successive differences, ratios, and, if needed, differences of ratios, until a constant row emerges. Once a constant row is found, the degree of the polynomial or the order of the recursion is known, and the next term can be projected by extending the difference table. Practising with deliberately noisy series — where one term is deliberately altered — trains the aspirant to verify the rule against every given element before committing to an answer.

Alphabet Positional Mapping with Modulo Arithmetic

Letter series become tractable when each alphabet is replaced by its positional index (A=1 … Z=26). The resulting numeric sequence can then be analysed with the same difference, ratio, and recursion tools used for number series. A frequent twist is the use of modulo‑26 arithmetic to model circular shifts: a forward shift of +3 from X (24) yields A (1) because (24+3) mod 26 = 1. Examiners exploit this by presenting series that appear to “wrap” at Z, such as Y, B, E, H, … where the underlying numeric pattern is +3 with wrap‑around. Another common device is the reverse‑position mapping (A↔Z, B↔Y) which corresponds to the transformation 27−p. Series built on reverse positions often combine a constant forward shift on the reversed indices, producing patterns like Z, W, T, Q (positions 26, 23, 20, 17). When a series mixes forward and reverse mappings, the numeric representation will show an alternating addition/subtraction pattern that is invisible in the raw letters. Modulo arithmetic also handles multi‑step jumps: a pattern of +5, −2, +5, −2 on positions translates to a net +3 every two steps, but the intermediate terms must respect the 1‑26 bounds. Aspirants should practise converting a given letter series to numbers, applying modulo‑26 reduction after each operation, and then reconverting to letters. This eliminates guesswork and makes the detection of hidden periodicities systematic. A further nuance appears when the series uses a keyword‑based substitution (e.g., a Playfair‑style keyword alphabet). In such cases the positional mapping is not the natural A=1 order but a permuted order; the series then behaves like a number series on the permuted indices. Recognising a keyword alphabet usually requires spotting a non‑standard frequency of certain letters or a given clue word in the question stem.

Technique: Modulo‑26 Difference Table
Find the next term in the letter series: K, N, Q, T, ?
Convert to positions: K=11, N=14, Q=17, T=20. First differences: +3, +3, +3. The constant difference indicates a simple arithmetic progression with step 3. Next position = 20+3 = 23. 23 corresponds to W. No wrap‑around occurs because 23 ≤ 26. Answer: W.
Exam‑Day Strategy
When a letter series looks irregular, immediately write down the positional numbers and compute first differences. If differences are not constant, compute second differences. Apply modulo‑26 only when a term exceeds 26 or falls below 1. This mechanical approach turns a visual puzzle into a numeric one, reducing careless errors under time pressure.

Compound Analogy Structures

Analogies in reasoning tests are rarely limited to a single semantic relation. A compound analogy stacks two or more distinct relations between the pair terms, and the candidate must replicate the entire stack on the target side. For instance, “Doctor : Hospital :: Teacher : School” encodes a worker‑workplace relation, but “Doctor : Stethoscope :: Teacher : Chalk” encodes a worker‑tool relation. When a question presents “Doctor : Hospital :: Stethoscope : ?”, the first pair gives worker‑workplace, the second gives tool‑user, and the required answer must be the workplace of the tool’s user, i.e., “Hospital”. Examiners also combine hierarchical relations: “Mammal : Dog :: Vehicle : Car” (class‑subclass) together with “Dog : Bark :: Car : Horn” (entity‑characteristic sound). A compound item may read “Mammal : Dog :: Bark : ?” expecting “Horn” because the first analogy establishes the subclass link and the second establishes the characteristic link, so the missing term must be the characteristic of the vehicle subclass. Another layer appears with part‑whole and material relations: “Wheel : Car :: Petal : Flower” (part‑whole) combined with “Car : Metal :: Flower : Organic” (object‑material). The test may ask “Wheel : Metal :: Petal : ?” where the answer is “Organic”. Solving these requires deconstructing the given analogy into its atomic relations, labeling each (e.g., R1 = worker‑workplace, R2 = tool‑user), and then applying the same sequence of relations to the new pair. A systematic approach: write the two given pairs side by side, identify all plausible relations for each, find the intersection of relation types that fits both pairs, and then project that relation chain onto the incomplete pair. Practising with a taxonomy of relation types — synonym, antonym, cause‑effect, degree, function, part‑whole, material, symbol, user‑tool, habitat, classification — builds the mental library needed to spot compound structures quickly.

Hybrid Alpha‑Numeric Series with Alternating Operations

Hybrid series interleave letters and numbers, often with two independent sub‑series running in parallel. A classic pattern is “A2, C4, E6, G8, ?” where letters advance by +2 (A→C→E→G) and numbers advance by +2 (2→4→6→8). The next term is I10. Examiners increase difficulty by making the two sub‑series follow different rules: letters may follow a quadratic index (1, 4, 9, 16 → A, D, I, P) while numbers follow a factorial growth (1, 2, 6, 24). The combined series would read “A1, D2, I6, P24, ?” and the next term would be “Y120” (25th letter Y, 5! = 120). Alternating operations within a single stream also appear: “B3, E6, H12, K24, ?” where letters +3 each step, numbers ×2 each step. The next term is N48. A more deceptive variant applies an operation to the pair as a whole, such as “A1, C3, F6, J10, ?” where the letter index and the number are identical (1, 3, 6, 10 are triangular numbers). The letter positions are the same triangular numbers mapped to letters (A=1, C=3, F=6, J=10). The next triangular number is 15 → O, giving “O15”. Recognising such cross‑dependency requires checking whether the numeric component equals the positional index of the letter, or whether a mathematical function (square, cube, prime) links them. A reliable solving routine: separate the series into letter‑only and number‑only streams, analyse each for independent patterns, then test for cross‑relations (equality, sum, product, modular equivalence). If a cross‑relation holds for all given terms, it is the governing rule. Time‑saving tip: write the positional indices of letters above the series; this visual alignment often reveals the hidden arithmetic instantly.

Pattern Recognition in Multi‑Dimensional Series

Advanced reasoning papers occasionally present series that evolve in two or three dimensions simultaneously, such as a grid of letters where each row and each column follows its own rule, or a three‑term tuple (letter, number, symbol) where each coordinate follows a distinct progression. Consider a 3×3 matrix where the first row is A1@, B2#, C3$; the second row D4%, E5&, F6*; the third row G7(, H8), I9?. The letter coordinate moves sequentially across the alphabet (A‑I), the numeric coordinate increments by 1 across each row and resets per row, and the symbol coordinate cycles through a fixed set @, #, $, %, &, *, (, ), ?. The missing term is I9). Solving such puzzles demands isolating each dimension: write the sequence of letters reading row‑wise, then column‑wise; do the same for numbers and symbols. Often the row‑wise pattern is simple (linear) while the column‑wise pattern is more complex (e.g., column 1 letters A, D, G are +3 each). The examiner may ask for the next element in a flattened series derived by reading the matrix in a spiral or diagonal order. In those cases, reconstruct the matrix, identify the traversal order, and then apply the per‑dimension rules to project the next tuple. Another multi‑dimensional form is a “triplet series” like (A, 1, +), (C, 4, ×), (E, 9, −), (G, 16, ÷) where the first coordinate advances +2, the second follows squares (1², 2², 3², 4²), and the third cycles through arithmetic operators. The next triplet is (I, 25, +). Mastery comes from practising the decomposition of each coordinate into its own one‑dimensional series, solving each independently, and then recombining. This modular method prevents the common error of forcing a single rule onto the whole tuple.

  • Assuming a single arithmetic rule governs a hybrid series without checking cross‑dimensional consistency.
  • Ignoring modulo‑26 wrap‑around when a letter shift passes Z, leading to an off‑by‑one error.
  • Treating a compound analogy as a simple one‑relation pair and missing the second relation layer.
  • Failing to separate interleaved sub‑series in alpha‑numeric patterns, causing a conflated difference table.
  • Overlooking the possibility of a keyword‑permuted alphabet when letter frequencies look non‑standard.
  • Projecting a row‑wise pattern onto a column‑wise request in matrix‑type series without verifying the traversal order.
  • Neglecting to test the derived rule against every given term before selecting the answer.

Recursive Polynomial Generation in Number Series

Number series that follow polynomial progressions of degree higher than two appear frequently in advanced reasoning papers. A quadratic series produces constant second differences; a cubic series yields constant third differences, and so on. The mechanism is to compute successive difference tables until a row of identical values emerges. That constant row reveals the degree of the governing polynomial. For instance, the series 2, 9, 28, 65, 126 shows first differences 7, 19, 37, 61; second differences 12, 18, 24; third differences 6, 6. The constant third difference of 6 indicates a cubic polynomial of the form an³ + bn² + cn + d. Solving the system for the first four terms yields a = 1, b = 0, c = 0, d = 1, giving the rule n³ + 1. The next term (n = 6) is 6³ + 1 = 217. Examiners often hide the polynomial by interleaving a simple arithmetic progression, e.g., 3, 10, 29, 66, 127, ? where each term equals n³ + 2. Recognising the cubic signature through third‑difference stability allows rapid projection without solving simultaneous equations each time. A practical shortcut: write the terms with their index n starting at 1, compute differences up to the third level, and if the third level stabilises, apply the formula (constant/6)n³ + lower‑order terms derived from the first few entries. This method also works for quartic series where the fourth difference becomes constant, though such cases are rarer and usually limited to the first five terms.

Cyclic Alphabet Shifts with Variable Modulo

Alphabet series frequently employ a shifting pattern that changes its step size according to a secondary numeric sequence. The core idea is to treat the alphabet as a cyclic group of order 26 and apply a modulo‑26 addition where the addend itself follows a recognizable series such as primes, squares, or Fibonacci numbers. For example, the series A, C, F, J, O, ? uses shifts of +2, +3, +4, +5 (triangular increments). Mapping letters to positions 1‑26, the positions become 1, 3, 6, 10, 15; the next triangular number is 21 → U. A more deceptive variant varies the modulo base: after reaching Z the series may continue with a different alphabet ordering, such as a keyword‑permuted alphabet where the effective modulo remains 26 but the mapping of indices to letters changes. In such cases, the solver must first reconstruct the permuted alphabet (e.g., keyword “REASON” yields R,E,A,S,O,N,B,C,D,F,G,H,I,J,K,L,M,P,Q,T,U,V,W,X,Y,Z) and then apply the same shift logic on the new indices. Another layer appears when the shift direction alternates: forward by the nth prime, backward by the nth square. The series B, E, A, J, ? demonstrates forward +3 (prime 3), backward −4 (square 4), forward +5 (prime 5), backward −9 (square 9). Tracking both the magnitude sequence and the sign pattern is essential. A reliable routine: write the positional indices, annotate each step with the applied shift and its source series, verify consistency across all given terms, then project the next shift from the source series and apply it modulo 26 on the current index.

Worked Example: Variable Modulo Alphabet Series
Find the next term in the series: M, Q, W, B, ? (Assume standard A=1…Z=26 alphabet.)
Map letters to positions: M=13, Q=17, W=23, B=2. Compute forward shifts modulo 26: 13→17 = +4, 17→23 = +6, 23→2 = +5 (since 23+5=28≡2). The shift magnitudes are 4, 6, 5. Observe the pattern of shift magnitudes: 4, 6, 5. These correspond to the sequence of composite numbers 4, 6, 8? but 5 breaks it. Check alternative: differences of shifts: +2, −1. Not stable. Consider the source series of shifts as the sequence of prime gaps: gaps between consecutive primes 2,3,5,7,11,13… are 1,2,2,4,2… not matching. Re‑examine the letter positions: 13, 17, 23, 2. The increments 4, 6, 5 could be the sequence of triangular numbers 1,3,6,10… offset? Actually 4,6,5 are the first three terms of the series “number of letters in month names” (Jan=3, Feb=3, Mar=3…) no. Simpler: the shifts follow the pattern +4, +6, +5, +7, +6, +8… alternating +2 then −1. If the pattern continues, next shift = +7 (since after +5 we add +2). Apply +7 to current index 2 → 9 → I. Verify: 2+7=9 (I). The series would be M, Q, W, B, I. This fits an alternating increment of +2, −1 on the shift magnitude. Therefore the missing term is I.
Exam‑Day Strategy for Variable Modulo Series
When a letter series shows irregular jumps, immediately write the positional indices and compute the signed shifts modulo 26. List the shift magnitudes in a separate column; then search for a known numeric pattern (primes, squares, triangular, composite, Fibonacci) in that column. If the shift magnitudes themselves form a recognizable series, project the next magnitude and apply it to the last index. This two‑layer decomposition (letter → index → shift → source series) prevents the common error of forcing a single arithmetic progression on the letters directly.

Nested Analogy Chains Across Multiple Relations

Compound analogies often require the solver to identify a chain of two or three distinct relations linking the first pair, then replicate the entire chain on the second pair. A typical structure is “A : B :: C : D” where A→B involves a synonym shift followed by a degree intensification, while C→D must undergo the same two‑step transformation in the same order. For instance, “Whisper : Shout :: Drizzle : Downpour” uses the relation chain (antonym of intensity low→high) applied twice. More complex chains combine different relation types: “Author : Manuscript :: Architect : Blueprint” follows (creator : product) then (product : representation). The examiner may embed a third relation such as “Manuscript : Edit :: Blueprint : Revise” adding a process step. To solve, first enumerate all plausible relations for the given pair using a taxonomy: synonym, antonym, cause‑effect, degree, function, part‑whole, material, symbol, user‑tool, habitat, classification. Then test each candidate chain against the second pair; the correct chain will satisfy every link without ambiguity. A practical method: write the first pair, draw arrows for each relation discovered, label them R1, R2, …; then apply R1 to the third term to generate an intermediate, apply R2 to that intermediate, and check whether the result matches the fourth term or its options. If multiple chains fit, look for the one with the fewest relation types (Occam’s razor) or the one that uses a relation explicitly hinted by the vocabulary (e.g., “seed : plant” suggests growth). Practising with a bank of 30‑40 predefined chains builds the mental library needed to spot nested structures within 30 seconds.

Matrix‑Based Alpha‑Numeric Series with Diagonal Traversal

Advanced papers present a two‑dimensional grid where each cell contains a composite token (letter‑number‑symbol). The series is read not row‑wise but along a diagonal, spiral, or knight‑move path. The solver must first reconstruct the matrix dimensions from the number of given terms, then identify the traversal rule. For example, a 4×4 matrix filled row‑wise with tokens A1@, B2#, C3$, D4%, E5&, F6*, G7(, H8), I9?, J10!, K11@, L12#, M13$, N14%, O15&, P16* yields a diagonal‑down‑right traversal starting at top‑left: A1@, B2#, C3$, D4%, then next diagonal E5&, F6*, G7(, H8), etc. The missing term may be the 10th element along that path. Isolating each coordinate (letter, number, symbol) across the traversal reveals independent one‑dimensional series: letters follow sequential alphabet, numbers increment by 1, symbols cycle through a fixed set of eight. The next token after P16* on the main diagonal would be the 17th element, but the matrix ends at 16; thus the series may wrap to the next diagonal starting at the second row first column. A systematic approach: (1) determine matrix order (square or rectangular) from term count; (2) write coordinates (row, column) for each given term assuming a filling order (row‑major, column‑major, or diagonal‑major); (3) deduce the traversal by comparing the sequence of coordinates; (4) project the next coordinate; (5) compute each component using its own series. This modular decomposition prevents the error of forcing a single rule on the composite token.

Conditional Rule Switching in Hybrid Series

Hybrid series sometimes change their governing rule after a specific condition is met, such as when a term exceeds a threshold, when a prime appears, or when the parity of the index flips. For instance, the series 2, 4, 8, 16, 31, 62, ? follows doubling until the term surpasses 30, after which the rule switches to “multiply by 2 then subtract 1”. The first four terms are 2×2ⁿ; the fifth term 31 = 16×2 − 1; sixth 62 = 31×2; seventh would be 62×2 − 1 = 123. Recognising the switch point requires monitoring a property of the current term (value > 30) rather than the index alone. Another variant uses the digital root: 5, 10, 20, 40, 80, 160, 320, 640, 1280, 2560, 5120, 10240, 20480, 40960, 81920, 163840, 327680, 655360, 1310720, 2621440, 5242880, 10485760, 20971520, 41943040, 83886080, 167772160, 335544320, 671088640, 1342177280, 2684354560, 5368709120, 10737418240, 21474836480, 42949672960, 85899345920, 171798691840, 343597383680, 687194767360, 1374389534720, 2748779069440, 5497558138880, 10995116277760, 21990232555520, 43980465111040, 87960930222080, 175921860444160, 351843720888320, 703687441776640, 1407374883553280, 2814749767106560, 5629499534213120, 11258999068426240, 22517998136852480, 45035996273704960, 90071992547409920, 180143985094819840, 360287970189639680, 720575940379279360, 1441151880758558720, 2882303761517117440, 5764607523034234880, 11529215046068469760, 23058430092136939520, 46116860184273879040, 92233720368547758080. The rule is simply ×2 throughout; no switch. But if a condition like “when the term contains digit 7” triggered a subtraction, the pattern would break. The solver must test each term for the condition, note the first occurrence, and then apply the new rule thereafter. A reliable checklist: (a) compute the default rule (e.g., ×2); (b) scan terms sequentially for the trigger; (c) at the trigger, apply the alternative rule to generate the next term; (d) continue with the alternative rule unless another trigger appears. This conditional awareness separates high‑scoring candidates from those who assume a single static rule.

  • Assuming a single polynomial degree without verifying higher‑order differences.
  • Ignoring modulo‑26 wrap‑around when a variable shift passes Z, causing off‑by‑one errors.
  • Treating a nested analogy as a single‑relation pair and missing the second or third relation layer.
  • Forcing a row‑wise reading on a matrix series that follows a diagonal or spiral traversal.
  • Overlooking a conditional rule switch because the trigger appears only once in the given terms.
  • Conflating the shift magnitude series with the letter position series in cyclic alphabet problems.
  • Neglecting to test the derived composite rule against every given term before selecting the answer.

Practice Questions

← Data Sufficiency ← Course Home