The chart and table items of the GMAT Data Insights section are where most candidates misallocate time, because the visual looks approachable and the prompt looks like a single short sentence. In practice these items reward a specific reading sequence: parse the prompt first, identify what kind of answer the test-maker is asking for, then walk back into the chart or table to extract only the values that matter. This article is built around that sequence. It covers the four prompt patterns that show up again and again in Graphics Interpretation and Table Analysis (and in the chart and table tabs of Multi-Source Reasoning), how scoring treats these items inside Data Insights, the concrete sub-skills that drive correct answers, and a preparation plan for a realistic six-to-eight-week window. The aim is for a candidate to leave the page with a method they can apply on the next 20 practice items, not a vague sense that the section is solvable.
Where chart and table items sit inside the GMAT Data Insights section
Data Insights is the 45-minute section of the current GMAT, with 20 questions drawn from five question types: Data Sufficiency, Multi-Source Reasoning, Table Analysis, Graphics Interpretation and Two-Part Analysis. There is no separate "Data Interpretation" question type; candidates who meet that label in prep materials are usually looking at Graphics Interpretation or Table Analysis items. Graphics Interpretation presents a graph or other image and asks you to complete statements about it by choosing from drop-down menus. Table Analysis presents a sortable table and asks you to judge a set of statements, each with a two-option response such as true/false or yes/no. Multi-Source Reasoning spreads its information across two or three tabs, some of which are charts or tables.
That structural picture is the first thing a tutor should press on. In Graphics Interpretation, the trap is misreading axis units, scale or a legend. In Table Analysis, the trap is something else: judging a statement that is not quite the one written, for example a statement about a share when the table gives totals, or about one year when the candidate has sorted on another. The reason this happens so often is that both item types put plausible numbers in front of the candidate, and candidates who read quickly enough to absorb the visual frequently skim the prompt and confirm a number that answers a different question. The three GMAT sections count equally towards the total score, and inside Data Insights every question type feeds the same section score.
For pacing within the section, the 45-minute window divided across 20 questions gives an average of 2 minutes 15 seconds per question. Chart and table items with a short prompt can often be finished faster than that average by a well-prepared candidate, which leaves time for the heavier Multi-Source Reasoning sets and Two-Part Analysis items. Candidates who ignore the prompt and dive into the visual routinely spend far longer on a single chart item, which then forces rushed work elsewhere in the section.
The four prompt patterns that frame every chart and table item
Prompts on chart and table items look varied on the surface, but they fall into four recurring patterns. A serious preparation plan treats these patterns as templates, because once a candidate can name the pattern, the rest of the work is largely mechanical. The four patterns are direct value, comparison, trend or change, and conditional or weighted. Each one demands a different way of reading the visual, and each one has a characteristic trap.
The first pattern, direct value, asks the candidate to read a single number from the chart or table. The statement or drop-down usually names one category and one period. The visual's role is to act as a lookup table. Most candidates handle this pattern well, but the trap is mis-scaling: confusing the right axis with the left axis, reading a stacked segment as a total, or pulling a value from a different year than the one specified. The defence is to read the prompt's reference labels, typically a category name and a time period, and find those exact labels on the visual before reading any number.
The second pattern, comparison, asks which of two entities is larger, smaller, faster, or higher in some metric. The prompt explicitly names both entities. The visual's role is to support a side-by-side read. The trap is choosing an entity that the visual does show but for a different period or category. The defence is to confirm both entity labels are present and that the comparison the prompt asks for is the comparison the visual supports. If a candidate cannot immediately see both entities in the form the prompt requests, the item is likely a trend or weighted prompt instead, and the candidate has misclassified the pattern.
The third pattern, trend or change, asks about a direction, a magnitude of change, or a rate. The prompt often includes a phrase like compared with, between X and Y, over the period, or at the highest point. The visual's role is to support a sequence read across categories or time. The trap is reading a single endpoint as the answer when the prompt asks about a movement. The defence is to identify the start point and the end point, then read both, and only then decide on the direction and the magnitude.
The fourth pattern, conditional or weighted, asks the candidate to apply a condition stated in the prompt to a value read from the visual. A typical prompt names a subset, a threshold, or a multiplier, and asks for a value or a judgement that respects that condition. The visual's role is to supply the raw data; the candidate must do the arithmetic, using the on-screen calculator where it saves time. The trap is forgetting the condition once inside the chart or table, especially when the arithmetic is trivial. The defence is to write the condition on the scratch pad before reading any number, so the condition is visible throughout the calculation.
Quick pattern-recognition checklist for the first 30 seconds
- Read the prompt first and underline the question being asked, not the numbers in the prompt.
- Name the pattern: direct value, comparison, trend or change, or conditional or weighted.
- Identify the labels the prompt references: entity, period, category, subset.
- Locate those exact labels on the chart or table before extracting any value.
- Decide on the reading strategy that the pattern requires before looking at the numbers.
How scoring rewards the prompt-first reading method
The GMAT reports a Data Insights section score on a 60 to 90 scale, and the section counts equally with Quantitative Reasoning and Verbal Reasoning towards the 205 to 805 total. The score reflects both how many questions you answer correctly and how difficult they were, because the section adapts question by question: after each answer, the computer re-estimates your level and uses that estimate to pick the next question. There is no separate sub-score for Graphics Interpretation or Table Analysis; the section score is what admissions committees see.
The scoring implication of the prompt-first method is that candidates who classify the pattern quickly and extract only the values the prompt needs leave more time for the heavier items and lose fewer points to misread answers. The reason is straightforward: the visual always contains more information than the prompt requires, and the extra information is the source of the most common wrong answers.
This is also why timed practice matters more than review of explanations for these items. A candidate who gets an item wrong because they misread the prompt and confirmed the right number for the wrong question will not learn much from reading the explanation, because the explanation will state the right number and the candidate will recognise it. The actual error is a sequencing error, and the way to fix it is to slow down the first read of the prompt and speed up the chart read. A preparation plan should therefore include repeated sessions of timed, prompt-first practice, followed by a careful review of the items where the wrong answer was the right number for a different question.
Common pitfalls and how to avoid them on chart and table items
- Reading the visual before the prompt. The fix is a strict rule: the chart or table is not touched until the prompt's question, labels and pattern are written down on the scratch pad.
- Answering a different question than the one asked. The fix is to restate each statement in your own words before judging it, so the test-maker's wording cannot be skimmed past.
- Misreading stacked or grouped chart segments. The fix is to read the legend first and to confirm whether the values on the axis are totals or components, then to read segments by following the legend.
- Forgetting a condition stated in the prompt. The fix is to underline the condition and keep it on the scratch pad through the calculation.
- Spending far more than the average on a single item. GMAC advises answering every question, even with an educated guess, so the fix is to enter your best answer, bookmark the item and keep going; at the end of the section you can review bookmarked items and change up to three answers.
Reading the visual with the prompt already classified
Once the prompt is classified, the read becomes a targeted extraction rather than a scan. The four patterns each call for a slightly different strategy, and the differences are worth rehearsing in practice. For direct value, locate the label and read across to the axis or column. For comparison, locate both labels and read both against the same axis or column, then compare. For trend or change, identify the start and end labels, read both, and compute the change. For conditional or weighted, read the condition, find the labels it references, read the relevant values, and apply the arithmetic on the scratch pad or the on-screen calculator rather than mentally.
A useful drill is to take a set of 10 practice items, classify each prompt by pattern without looking at the visual, and only then look at it. This drill exposes how often a candidate's first instinct is shaped by a pattern they have not consciously identified, and it forces the classification to happen at the prompt stage. After a few sessions, the prompt read becomes automatic, and the chart read compresses because the candidate knows what they are looking for before they look.
The habit plays out a little differently by item type. In Graphics Interpretation, the prompt is the statement with its drop-down blanks, and the work is to read each blank as a question before looking at the graph. In Table Analysis, the prompt is each statement, and the first decision is which column to sort by; sorting before reading the statement is the most common time leak. In Multi-Source Reasoning, the same pattern-naming step tells you which tab to open first.
A six-to-eight-week preparation plan for chart and table items
A focused plan for chart and table items should run in parallel with broader Data Insights preparation, because the reading skills transfer across the five question types but the pattern recognition needs its own practice. Week one should be diagnostic: a timed 45-minute Data Insights section under test conditions, with the candidate tracking which items were Graphics Interpretation or Table Analysis, how long each took, and whether the wrong answer was the right number for the wrong question or a true chart-read error. The diagnostic tells the candidate whether their bottleneck is pattern classification, chart reading, or pacing.
Weeks two and three should be pattern drill. The candidate takes sets of 15 to 20 chart and table items, classifies each prompt by pattern before looking at the visual, and reviews the items where the classification was correct but the answer was wrong. Items where the classification was wrong are re-attempted at the end of the week with the pattern label visible, to confirm the candidate can recognise the pattern when prompted. By the end of week three, naming the pattern should take only a few seconds.
Weeks four and five should be full Data Insights sections under timed conditions, with the candidate reviewing every chart and table item regardless of correctness. The review focuses on three questions: was the pattern correctly identified, was the read aligned with the pattern, and was the pacing inside the section average. Log the answers for every item, because patterns of error across items are easier to see in a log than in a memory.
Week six should be mixed review, alternating between chart and table drill sets, full Data Insights sections, and the heavier question types such as Multi-Source Reasoning. The aim is to consolidate the prompt-first habit under section-level pressure. Weeks seven and eight should be taper: full-length practice tests with all three sections, and a final review of the log to confirm that the most common error has shifted from misread prompts to a small residual class of chart-read errors that need a different fix, often axis-unit or legend care.
How chart and table items compare with the other Data Insights question types
Graphics Interpretation and Table Analysis are the two Data Insights types built around a single visual. Graphics Interpretation uses a graph or image and drop-down completions. Table Analysis uses a sortable table and a set of two-option statements. Multi-Source Reasoning uses two or three tabs of text, tables and charts, with questions that ask the candidate to combine information across the tabs. Two-Part Analysis asks for two linked answers chosen from a shared set of options. Data Sufficiency asks whether two statements are sufficient to answer a question, with a fixed set of five answer choices.
The payoff from prompt-first reading is highest on Graphics Interpretation, Table Analysis and Data Sufficiency, because all three punish the candidate who reads the data first and the question second. On Multi-Source Reasoning, the sequencing is a triage of which tab to read first. Candidates who build a strong prompt-first habit on chart and table items often find the same habit transfers to Data Sufficiency, where the prompt is the question and the two statements are the data.
| Question type | Visual | Response format | Primary sequencing habit |
|---|---|---|---|
| Graphics Interpretation | A graph or other image | Drop-down menus that complete statements | Read each blank as a question, classify the pattern, then read the graph |
| Table Analysis | Sortable multi-column table | Several statements, each with a two-option response | Read the statement, decide which column to sort, then read off the value |
| Multi-Source Reasoning | Two or three tabs of text, tables and charts | Multiple-choice or two-option statement questions | Triage the tabs, then read the one the prompt references |
| Two-Part Analysis | Short scenario with a shared option list | Two linked selections | Name the relationship between the two parts before testing options |
| Data Sufficiency | Question with two statements | Fixed five answer choices | Read the question, evaluate statement 1 alone, then statement 2, then both |
Drills that produce measurable improvement on chart and table items
The most efficient drill is a 20-item timed set at section pace, with the prompt-first method enforced by a visible rule: the visual is not touched until the prompt has been read and the pattern has been written on the scratch pad. After the set, review the items where the wrong answer was the right number for the wrong question, and re-attempt those items the next day to confirm that the pattern classification now sticks. Over a few sessions, pattern misclassification should fall, and the residual errors should be chart-read errors that respond to a different drill.
The second drill is the reverse set, where the candidate looks at the chart or table first and writes down what it could plausibly ask, then compares with the actual prompt. This drill is for candidates whose bottleneck is over-reliance on the visual, and it surfaces the gap between what the data could support and what the prompt actually asks. Keep it short, five to seven items, and run it once a week rather than daily, because the cognitive load is high.
The third drill is the arithmetic drill, focused on the conditional or weighted pattern. Take 10 items where the prompt states a condition or a multiplier, and practise writing the condition on the scratch pad before reading the data. The aim is to make the scratch-pad step automatic, so the condition stays visible during the arithmetic and cannot be forgotten.
Putting the method together on test day
On test day, the prompt-first method collapses into a short, rehearsed sequence: read the prompt, name the pattern, identify the labels the prompt references, locate those labels on the chart or table, extract the values the pattern needs, compute on the scratch pad or calculator, and choose the answer that matches the question the prompt asked. The sequence keeps chart and table items inside the section's average time and leaves room for the heavier question types. Candidates who have rehearsed it in timed practice should not need to think about it on test day; the sequence is the habit.
The single most important tactical point, in my experience tutoring candidates on the GMAT, is that wrong answers on chart and table items are driven by sequencing errors more often than by chart-reading errors. Candidates who already read charts well are often the ones who struggle most, because their chart skill tempts them into the visual first. Reversing the sequence, even for candidates with strong visual reasoning, is the change that pays off fastest. A diagnostic that separates sequencing errors from chart-read errors is the first step, and a six-to-eight-week plan that drills the sequence under timed pressure is the rest.