HARCO Bank Clerk common mistakes
HARCO Bank Clerk · Understand the traps and how to avoid them.
In Data Interpretation sets, reading the graph or table once and then computing every sub-question from memory instead of re-checking the exact row, column, or legend label for each question, causing answers built on the wrong series or a misread axis scale.
Time pressure pushes candidates to 'save' the 5-10 seconds of re-reading the exhibit, and DI sets are deliberately built with similar-looking categories (e.g., two years, two banks, gross vs net) so a single misread propagates through all five questions.
What to do instead
Adopt a fixed per-question routine: restate the exact row/column/unit in one short phrase before calculating, and mark units (lakh/crore, %, thousands) at the top of your rough sheet. Practise 10 sets where you write the label reference next to each calculation so the habit becomes automatic.
Approximating in simplification and approximation questions but then selecting the option that matches the exact computed value rather than the nearest option, or vice versa, without checking how far the options are spaced.
Candidates learn 'approximate to save time' and 'compute exactly to be safe' as separate rules and never decide which regime applies before starting, so they either waste 90 seconds on exact long division or round so aggressively that two options become indistinguishable.
What to do instead
Before solving, glance at option spacing: if options differ by more than about 2%, round inputs to two significant figures and answer fast; if they are within 1%, compute exactly. Drill 30 questions tagging each as 'round' or 'exact' first, then solving, to build the judgement.
In reasoning puzzles and seating arrangement sets, starting to fill the arrangement from the first clue without first identifying the fixed anchor (a definite position, a definite pair, or a definite block) and without listing which clues are conditional.
The first clue read feels like the natural starting point, and under time pressure candidates begin placing people immediately; conditional clues ('if A sits adjacent to B') then get treated as fixed facts, producing a plausible but wrong arrangement that consumes the whole set's time.
What to do instead
Spend the first 30 seconds classifying each clue as definite or conditional and drawing the skeleton (number of seats, facing direction, categories). Only start placing from definite clues. If two arrangements survive after all definite clues, test the conditional clues against both rather than guessing.
In English Language, answering reading-comprehension and para-jumble questions from general knowledge or from the topic of the passage rather than locating the specific sentence or logical connector that decides the answer.
Fluent readers skim and rely on gist, which works for casual reading but fails on inference, tone, and 'which of the following is NOT stated' questions where the exam deliberately places a near-true distractor. Para-jumbles are attempted by 'what sounds right' rather than by opening-closing pairs and pronoun/connector links.
What to do instead
For RC, answer only after underlining the line that supports the chosen option, and for NOT/EXCEPT questions eliminate the three options that are explicitly stated. For para-jumbles, find the opener (no pronoun, no definite article referring back), then chain mandatory pairs using pronouns and connectors before checking the full sequence.
In Banking and Financial Awareness and General Awareness, revising only static facts (history of banks, capitals, founders) and skipping current monetary-policy numbers, recent scheme names, and RBI regulatory changes from the last 6-12 months.
Static content is easy to find in compiled PDFs and feels productive, while current-affairs revision requires weekly discipline; candidates also assume clerk-level awareness is 'basic' and will not ask about repo-rate changes, CRR/SLR levels, or newly launched schemes.
What to do instead
Maintain a one-page rolling sheet updated weekly with: current repo/reverse repo/CRR/SLR/MSF rates, latest CPI and GDP prints, RBI governor and deputy directions, newly notified schemes with launch purpose, and recent banking appointments. Revise that sheet every third day; test yourself with 20 current-banking MCQs weekly.
In Computer Knowledge and Digital Banking, memorising shortcut keys and file extensions but being unable to answer scenario questions about UPI transaction flows, NEFT vs RTGS vs IMPS settlement timing, or what a phishing/OTP-fraud attempt looks like.
Preparation material for this section is dominated by static IT facts, while the actual paper increasingly asks applied digital-banking questions; candidates never map the terms to a real transaction, so they cannot distinguish, for example, RTGS real-time gross settlement from IMPS immediate retail transfer.
What to do instead
Build a comparison table for NEFT, RTGS, IMPS, UPI and AePS covering: settlement basis, timing, minimum/maximum limits, and who can use it. Add a short list of common fraud patterns (vishing, OTP sharing, QR-code collect-request scams) with the correct safe behaviour for each. Revise the table weekly.
In Clerical Aptitude, copying the entire original string and the entire comparison string character by character instead of scanning for the first point of difference, and not deciding in advance whether the task is 'same/different' or 'count the differences'.
The section rewards speed over depth; candidates who read both strings fully spend 3-4 times longer per item and run out of time, while those who rush without a fixed scanning method miss transpositions and look-alike characters (0/O, 1/l, rn/m).
What to do instead
Practise a left-to-right chunked scan in blocks of 3-4 characters, stopping at the first mismatch for same/different items. For counting-difference items, tick each mismatch position on the answer sheet. Time yourself on 50 items in 5 minutes and log which character pairs you miss most.
In Quantitative Aptitude word problems (time-work, trains/boats, mixtures, partnership), writing the equation directly in one line without defining the variables and units, then being unable to recover when the answer does not match any option.
Candidates have practised enough to recognise problem types but not enough to structure them; under exam pressure they skip the definition step, mix units (minutes vs hours, metres vs km), and have no audit trail to find the error, so they either abandon the question or re-solve from scratch.
What to do instead
Force a two-line setup: define each variable with its unit, then write the relation. Convert all units to a single system before substituting. If the answer is absent from the options, re-check the unit conversion first, not the algebra. Do 20 problems where the setup lines are graded separately from the arithmetic.
In Reasoning Ability, attempting every question in strict paper order, sinking 3-4 minutes into one difficult puzzle or a lengthy input-output set, and thereby losing the easy inequality, direction-sense, and syllogism questions at the end.
Candidates treat the paper as a test of completeness rather than of selection; the first difficult set creates a sunk-cost trap, and the fear of leaving questions unattempted overrides the arithmetic of negative marking and time value.
What to do instead
Adopt a two-pass strategy: on pass one, do all standalone reasoning items (inequalities, syllogisms, coding, blood relations, direction) and skip any puzzle set you have not cracked in 90 seconds. Return to skipped sets in pass two with the remaining time. Rehearse this with full-length mocks and record how many marks the two-pass order gains.
In English Language, guessing options for fill-in-the-blank, error-spotting, and cloze items by 'which sounds natural' rather than by checking grammar rules (subject-verb agreement, preposition collocation, tense sequence) and without using the elimination of grammatically impossible options.
Everyday spoken English tolerates many constructions that the exam marks wrong, and candidates trust fluency over rules; they also read the sentence once and commit to the first option that fits the meaning, ignoring that two options may be semantically close but only one is grammatically correct.
What to do instead
For each blank, first eliminate options that are grammatically impossible regardless of meaning (wrong preposition, wrong tense form, wrong countability), then choose among survivors by meaning. Maintain an error log of every rule you miss (e.g., 'discuss about', 'married with', 'one of the + plural noun + singular verb') and re-test it weekly.