Aon gapChallenge: Deduce the Missing Shape Without Guessing

Aon gapChallenge preparation should teach you to prove the missing shape, not fill every blank or guess from visual balance. The central habit is to keep the row and column constraints separate, intersect their candidate sets, and use a helper cell only when it removes ambiguity at the target.

Aon's official preparation page describes gapChallenge as a six-minute deductive-reasoning assessment. Your invitation and on-screen instructions remain authoritative. The grids below are original practice exercises, not copied Aon questions, employer reports, or a prediction of your assessment.

For a related reasoning exercise, try Deduce position from logic clues. The transferable skill is explaining which constraint forces an answer; the interface and scoring are different.

What the official practice material establishes

Aon's gapChallenge practice PDF illustrates four-by-four and five-by-five grids. Shapes must not repeat within a row or column. Some cells are empty, and the task is to identify the shape at the question mark. Other empty cells can help the deduction, but completing the whole board is not the objective.

The PDF is an official practice resource, not proof that every live invitation uses exactly its examples or presentation. Its practice instructions should not be treated as permission to use paper, external tools, or assistance during a scored assessment. Confirm permitted aids and submission behavior in your own instructions.

This article narrows its claims to official mechanics and original deduction practice. Available candidate discussions did not establish two independent comparable completed-test reports for a current employer configuration. Consequently, it does not claim a universal question count, cutoff, adaptive scoring formula, or hiring sequence.

Do not confuse gapChallenge with an odd-one-out task. You are not looking for a shape that violates a visual pattern; you are finding a value consistent with explicit placement constraints. Likewise, a memory game asks you to retain information, while this exercise asks you to derive it from what remains visible.

Represent the constraints before scanning for an answer

For the original examples, use four symbols: C for circle, T for triangle, S for square, and X for cross. Each row and column contains every symbol exactly once when complete. A dot is an empty cell, not a fifth symbol. Rows are numbered from top to bottom; columns from left to right.

The candidate set for an empty cell contains symbols missing from its row and missing from its column. A symbol ruled out by either dimension cannot be placed there. This intersection is stronger than the statement “the row needs a circle,” because the row may also need other shapes.

For example, if a row is missing circle and square while the column is missing circle and triangle, only circle survives both constraints. You need not know where the row's square will go to prove that result. Local certainty at the target is sufficient.

Be explicit about the rules you are not using. These exercises have no extra two-by-two box restriction, no diagonal restriction, and no symmetry requirement. Recognizing a familiar Sudoku-like layout does not authorize importing Sudoku's additional constraints. An attractive pattern is not evidence unless the instructions make it a rule.

Exercise one: solve by immediate intersection

This original board has a target at row 1, column 1:

? T . X
S . . .
X . . .
. . . .

Row 1 already contains triangle and cross, so the target's row permits circle or square. Column 1 already contains square and cross, so the target's column permits circle or triangle. Their intersection is circle. Therefore the answer is C.

This is a proof rather than a preference. Square would repeat in column 1; triangle would repeat in row 1; cross would repeat in both. Circle is the only surviving symbol. You can also demonstrate feasibility with a completion, but you do not need to complete the board to select the target.

Avoid turning that deduction into a slow full-board workflow. Once the intersection contains exactly one symbol, stop exploring irrelevant blanks. Additional work can create unnecessary opportunities to transpose a row, forget the target position, or click the wrong option.

As a rehearsal, cover the explanation and state the two candidate sets aloud before naming the answer. If you cannot identify why each alternative is excluded, you may be recognizing the solution rather than reproducing the reasoning. Change the symbol names and repeat; the logic should survive that change.

Exercise two: use one helper, then stop

The next original board uses the same row and column rules:

? . S .
T . X .
. . . .
. C . S

At first, row 1 permits circle, triangle, or cross. Column 1 already contains triangle, so the target candidates are circle and cross. Neither can yet be selected from the target's immediate intersection alone.

Now inspect row 4. It contains circle and square, leaving triangle and cross for its two blanks. The blank at row 4, column 1 cannot be triangle because column 1 already contains triangle in row 2. That helper cell must therefore be cross.

Return to the target. Column 1 now contains triangle and cross; row 1 contains square. Circle is the only candidate that survives. The helper has done its job. Do not continue filling the remaining blanks merely because you have established momentum.

Step	Information used	Deduction
Inspect the target	Row 1 has S; column 1 has T	Target candidates are {C, X}.
Inspect helper (4,1)	Row 4 needs {T, X}; column 1 already has T	Helper is X.
Revisit target (1,1)	Column 1 now rules out T and X; row 1 rules out S	Target is C.

A unique target does not require a unique whole board

The second board has two complete valid arrangements. Both assign circle to the target, but some other cells differ:

Completion A    Completion B
C T S X         C X S T
T S X C         T S X C
S X C T         S T C X
X C T S         X C T S

Check each row and column against the four-symbol rule and compare the original givens. Both completions are valid. Their disagreement elsewhere does not weaken the target proof because they agree on the only cell the question asks you to identify.

This distinction helps when an unfinished board feels uncomfortable. An empty cell is not necessarily missing information you must recover. If every valid completion gives the target the same symbol, the target is forced even when the rest of the grid is not.

The bounded Python verifier below enumerates completions of this original four-by-four exercise. It is an offline preparation check, not a tool to use during a live assessment. The search chooses an empty cell with the fewest remaining candidates and explores only values consistent with its row and column.

symbols = set("CTSX")
board = [list(r) for r in ["..S.", "T.X.", "....", ".C.S"]]
solutions = []

def complete():
    choices = []
    for r in range(4):
        for c in range(4):
            if board[r][c] == ".":
                used = set(board[r]) | {board[k][c] for k in range(4)}
                options = symbols - used
                choices.append((len(options), r, c, options))
    if not choices:
        solutions.append(["".join(row) for row in board])
        return
    _, r, c, options = min(choices)
    for value in sorted(options):
        board[r][c] = value
        complete()
        board[r][c] = "."

complete()
assert len(solutions) == 2
assert {solution[0][0] for solution in solutions} == {"C"}
assert solutions[0] != solutions[1]

The code corroborates the exercise; it does not replace a concise human explanation. Its result is specific to this finite board, and it does not establish anything about proprietary item generation or scoring. For a larger reasoning problem, distinguish a proof about the target from an exhaustive search you cannot practically perform.

When the target still has two candidates

Do not guess immediately. Look for a nearly complete row or column that intersects the target's unresolved constraints. A forced helper elsewhere can remove one candidate. In the second exercise, the useful helper sits in the target's column and has a restrictive row; that is why it matters.

If no direct forced cell is visible, inspect where a missing symbol can go within a row or column. A symbol may have several empty locations but only one legal location after checking their other dimension. Explain that restriction rather than make an unsupported placement.

During untimed original practice, you can test a tentative assignment and look for contradiction. Label it as an assumption, keep deductions derived from it separate, and undo the entire branch if it fails. A contradiction might be a repeated symbol or an empty cell with no candidates. Do not silently convert a tentative value into a given.

Whether branching is a sensible live-test strategy depends on your instructions, remaining time, and interface. This article does not assume you can annotate, reset, skip, or return to a question. Practise learning those controls during the available explanation stage so that a logical plan does not depend on a nonexistent feature.

Diagnose the mistake before trying to become faster

A useful error log separates constraint errors from interaction errors. “Used only the row” calls for candidate-set practice. “Added a diagonal rule” calls for instruction discipline. “Solved the helper but never returned to the target” calls for a target-focused workflow. “Selected square while intending circle” calls for a final answer-to-option check.

For each missed original practice question, record the target coordinate, the mistaken candidate set, and the exact rule that should have removed the wrong answer. Re-solve a changed board rather than merely repeat the same one until its answer feels familiar.

Use short timed sessions only after the reasoning is accurate. Keep the practice clock separate from any claim about a real scoring system. Compare your own accuracy on similar exercises, but do not translate a home-made count into a percentile or employer pass threshold.

Aon's ability-test report guidance discusses interpreted assessment results and comparison context. It does not supply a public gapChallenge points-per-item formula. A self-made practice score therefore cannot establish the meaning of a real report.

Before submitting an answer, verify three things: you solved the question-mark cell, your chosen symbol violates neither its row nor its column, and the selected option matches that symbol. This check is brief because it reuses the deduction; it does not require a full-board reconstruction.

Practise the reasoning, not an alleged answer bank

Some add rules that gapChallenge's public examples do not have, so read each problem's own contract. They are not official Aon items or interface simulations.

Question	Transferable habit
Solve a Skyscraper puzzle efficiently	Propagate explicit constraints without importing unrelated rules.
Solve a logical reasoning puzzle	Separate a forced conclusion from a plausible arrangement.
Deduce position from logic clues	Explain which clue removes each alternative.
Solve Logic Puzzles: Algorithms and Proofs Required	Distinguish construction, exhaustive checking, and proof.
Find the Top-Left Submatrix Matching a Grid Pattern With Letter Variables	Track positions and symbol consistency carefully.

Start with Deduce position from logic clues, then practise the two original boards without their explanations. Your goal is a short, reproducible proof: target candidates, useful helper if needed, elimination, and final option check. Stop when the requested shape is established.

Sources and Further Reading

• Aon: Prepare for your online assessment
• Aon: Deductive reasoning gapChallenge practice tasks, PDF
• Aon Assessment support: Ability and skills test reports