Accenture Placement Papers 2026: Cognitive and Coding Guide

Accenture’s fresher selection has four gates, and the Cognitive Ability test is where the highest volume of candidates stop advancing.

This article covers the written and coding gates in detail: the Cognitive Ability test (Critical Reasoning, Abstract Reasoning, Verbal Ability), the Technical Assessment (pseudocode, CS fundamentals), and the Coding round. The spoken-English Communication Assessment is a separate test with its own AI-graded format; for that, see the Accenture Communication Assessment guide.

The two Accenture hiring tracks and who qualifies

Accenture’s campus recruitment runs on two distinct tracks, both starting from the same eligibility floor.

Eligibility (both tracks):

• Aggregate of 60% or equivalent CGPA throughout Class 10, Class 12, and all semesters of graduation
• No active backlogs at the time of application
• Any engineering branch: CSE, ECE, EEE, IT, AIDS, Mech, Civil — all qualify

Track	Role	CTC range	What differentiates selection
Standard	Associate Software Engineer (ASE)	Rs 4.5 to 6.5 LPA	Standard Cognitive, Technical, and Coding threshold
Premium	Advanced ASE (11A grade)	Rs 6.5 to 9.0 LPA	Higher Coding round score and deeper technical interview

CTC bands sourced from FACE Prep placement data, 2026 intake cycle.

The practical implication: if your coding performance is strong, the Advanced ASE track is worth targeting from the start. There is no separate application; Accenture’s scoring system places candidates into tracks based on their test results.

How the four selection rounds are sequenced

Accenture runs its selection in a fixed order. A candidate who does not clear one round does not proceed to the next.

Round	What it tests	Where this article covers it
1. Communication Assessment	Spoken English (AI-graded, 6 sections, 63 questions)	Dedicated article linked below
2. Cognitive Ability test	Critical Reasoning, Abstract Reasoning, Verbal Ability	H2 below
3. Technical Assessment	Pseudocode, CS fundamentals, networking, cloud	H2 below
4. Coding round	Algorithm and data-structure problems	H2 below
5. Technical and HR interview	Role fit, technical depth, HR questions	Linked guide below

The Communication Assessment runs first and is entirely audio-based: 6 sections, 63 questions, all graded by AI with no human evaluator in the room. A dedicated guide covers the full section pattern and preparation strategy for it: Accenture Communication Assessment guide.

The rest of this article focuses on rounds 2, 3, and 4.

Cognitive Ability test: section-by-section pattern

The Cognitive Ability test covers three sections, all multiple-choice. Questions move at a fixed pace; time management between sections matters as much as correctness. The section below covers the standard topic distribution, worked examples, and where students typically lose marks.

Section	Core topics	What the question type tests
Critical Reasoning	Time-speed-distance, percentages, blood relations, syllogisms, seating arrangements, data interpretation, coding-decoding	Problem-solving and deductive reasoning
Abstract Reasoning	Number series, letter series, odd-one-out, matrix patterns, pattern completion	Pattern identification without arithmetic
Verbal Ability	Reading comprehension, synonyms and antonyms, fill in the blanks, sentence correction, para-jumbles	Grammar, vocabulary, reading accuracy

Critical Reasoning: worked example

Problems from the time-speed-distance category appear regularly. This type of question rewards a unit-conversion habit more than any formula.

• Question: A train running at 72 km/hr crosses a bridge in 50 seconds. The train’s length is 200 metres. Find the length of the bridge.
• Step 1: Convert speed to m/s: 72 multiplied by (5/18) = 20 m/s
• Step 2: Total distance covered in 50 seconds: 20 multiplied by 50 = 1,000 metres
• Step 3: Length of bridge = total distance minus train length = 1,000 minus 200 = 800 metres

Blood-relations questions follow a different pattern: map the relationships as a diagram rather than tracking them in text. A typical question gives three or four relationships in sequence (“A is the father of B; B is the sister of C; D is the mother of C”) and asks how A and D are related. Drawing the tree takes five seconds and eliminates errors that come from trying to hold the chain in working memory.

Abstract Reasoning: worked example

Number series questions test whether you can identify the rule quickly, not whether you can do heavy arithmetic.

• Question: Find the missing number: 2, 6, 12, 20, ?
• Pattern: Each term equals n squared plus n, where n is the position. Position 1: 1 plus 1 = 2. Position 2: 4 plus 2 = 6. Position 3: 9 plus 3 = 12. Position 4: 16 plus 4 = 20. Position 5: 25 plus 5 = 30.

For odd-one-out questions using letter clusters (e.g., BDG, FHK, OQT, NPT), check the letter gaps first. BDG: gaps of 2, 3. FHK: gaps of 2, 3. OQT: gaps of 2, 3. NPT: gaps of 2, 4. NPT is the odd one out. The gap-counting method works faster than trying to assign meaning to the cluster.

Verbal Ability: worked example

Sentence correction questions test basic grammar rules, particularly perfect-tense verb forms.

• Question: Choose the grammatically correct sentence:
  • a) She has been went to the market.
  • b) She has gone to the market.
  • c) She had went to the market.
  • d) She has go to the market.
• Answer: b) “She has gone to the market.” The present perfect tense requires the past participle “gone,” not “went” (which is the simple past, not the past participle).

Reading comprehension passages in the Verbal section are typically 150 to 200 words. The questions test inference and tone, not just factual recall. Read the questions before the passage on the first pass. That directs your reading toward relevant sentences rather than full re-reading.

For an extended set of reasoning problems with full solutions across logical and non-verbal categories, the Accenture Logical Reasoning question bank covers additional pattern types.

Technical Assessment: pseudocode and CS fundamentals

The Technical Assessment follows the Cognitive section. It tests applied CS knowledge rather than abstract reasoning. Three areas consistently appear across campus recruitment cycles.

1. Pseudocode (highest weightage)

Pseudocode questions give you a short block of logic and ask for the output. The skill being tested is reading code without running it.

• Question: What does the following pseudocode output?

  X = 5
  Y = 10
  If (X + Y > 10)
      Print "Greater"
  Else
      Print "Smaller"

• Answer: Greater — because 5 plus 10 = 15, and 15 is greater than 10, so the condition is true.

The pseudocode section uses simple conditionals, loops, and variable assignments. The most common errors come from misreading operator precedence or mentally skipping an else branch that doesn’t execute.

2. CS fundamentals: database, networking, and OOP

These questions draw from core engineering subjects that all branches cover in their curriculum.

Database question example:

• Which SQL query fetches the highest salary from an employees table?
  • a) SELECT MAX(salary) FROM employees;
  • b) SELECT salary FROM employees WHERE salary = MAX();
  • Answer: a. The MAX() aggregate function is used inside SELECT, not in a WHERE clause directly.

Networking question example:

• What does HTTPS stand for?
  • a) HyperText Transfer Protocol Secure
  • b) High Transfer Text Protocol Secure
  • Answer: a. HyperText Transfer Protocol Secure.

OOP question example:

• Which statement about inheritance is correct?
  • a) A subclass can access all private members of its parent class.
  • b) A subclass inherits both data members and member functions of its parent class.
  • c) Multiple inheritance is supported in Java using classes.
  • Answer: b. A subclass inherits public and protected members. Private members are not directly accessible, and Java does not support multiple inheritance through classes (only through interfaces).

3. MS Office and common applications

A smaller portion covers Excel functions and standard office software.

• Which Excel function returns the highest value in a range?
  • a) MAX() b) MIN() c) SUM() d) AVERAGE()
  • Answer: a. MAX()

Coding round: problems, languages, and worked examples

The Coding Assessment runs after the Technical section. Candidates choose one language from C, C++, Java, Python, or .NET per problem. Most campus drive candidates choose Java or Python; both have strong standard library support for the string and array tasks that appear most often.

The standard ASE track has one to two coding problems at a difficulty level comparable to basic data structures and string operations. The Advanced ASE track increases problem complexity, with greater emphasis on edge-case handling and algorithmic thinking rather than brute-force solutions.

Solved example 1: password validation

This type of string validation problem appears frequently in entry-level assessments. The task is checking multiple conditions on a single input without missing any edge case.

Problem: Write a function that checks whether a string is a valid password. Rules:

• Minimum length of 4 characters
• Must contain at least one digit and at least one uppercase letter
• No spaces or ’/’ characters
• Must not start with a digit

Input: aA1_67 Expected output: 1 (valid)

def is_valid_password(s):
    if len(s) < 4:
        return 0
    if s[0].isdigit():
        return 0
    if ' ' in s or '/' in s:
        return 0
    has_digit = any(c.isdigit() for c in s)
    has_upper = any(c.isupper() for c in s)
    return 1 if (has_digit and has_upper) else 0

print(is_valid_password("aA1_67"))  # Output: 1

Verify: length 6 passes, starts with ‘a’ not a digit passes, no space or ’/’ passes, contains ‘1’ as a digit passes, contains ‘A’ as an uppercase letter passes. Output is 1. Correct.

Solved example 2: second largest element

This is a direct data-structure question: no recursion, no complex algorithm, but it requires handling edge cases like duplicate values.

Problem: Find the second largest distinct element in an array. Return -1 if no such element exists.

Input: [3, 1, 4, 1, 5, 9, 2, 6, 5, 3] Expected output: 6

def second_largest(arr):
    unique = list(set(arr))
    if len(unique) < 2:
        return -1
    unique.sort(reverse=True)
    return unique[1]

print(second_largest([3, 1, 4, 1, 5, 9, 2, 6, 5, 3]))  # Output: 6

The set() removes duplicates. After sorting descending: [9, 6, 5, 4, 3, 2, 1]. The second element is 6. Correct.

For the interview stage that follows both tracks, the Accenture Technical and HR interview questions guide covers what to expect and how interviewers frame technical questions for freshers.

AI skills and your Accenture application in 2026

The coding round makes one thing consistent across Accenture’s selection process: the test rewards candidates who produce working output, not just theoretical knowledge. That principle extends to the broader hiring context.

IT sector fresher hiring is projected to add 150,000-plus roles in FY26, with Accenture among the firms driving this expansion with an explicit GenAI focus (Economic Times, early 2025). Accenture’s Generative AI Scholars Program offers 40-plus hours of self-paced GenAI learning built on Stanford Online content through the LearnVantage platform, launched in 2024. That platform exists for one reason: help employees and freshers build AI-economy skills before the job requires them. For freshers targeting the Advanced ASE track, a demonstrable AI project at the HR stage carries the same weight as a clean coding-round submission: it shows the output, not just the intent.

For students building that layer now, the 2026 AI roadmap for Indian engineering students maps the learning sequence from LLM fundamentals through to a deployable project. TinkerLLM (Rs 299) offers a browser-based build environment where you can ship an LLM-powered tool without local machine setup.