GMAT Focus tables and spreadsheets

GMAT Focus tables and spreadsheets are the most visually imposing items in the Data Insights section, and they are the items candidates most often misread before the maths even starts. The question type presents a small data grid, a short business-style prompt, and three statements that must each be classified as true or false, with a final interpretive question that often asks which statement is correct or how many are true. The screen looks busy; the actual work is column logic, footnote discipline, and ruthless attention to units. A candidate who treats the table as decoration will burn four minutes and still pick the wrong statement. A candidate who treats the table as the entire question will finish with ninety seconds to spare. The aim of this article is to put you in the second group, with a working method for reading the prompt, mapping columns to claims, and handling the sortable versus fixed layout variants that the GMAT Focus edition uses.

What the GMAT Focus is actually testing on table analysis items

The first thing to internalise is what the question type is, and is not, measuring. It is not a calculation test in the way that problem solving is. Yes, you will divide one column by another, you will compute a difference, and occasionally you will need a quick percentage. But the arithmetic budget per statement is small, often a single division or a single subtraction. What the question really tests is your ability to scan a structured dataset, identify the correct row, isolate the correct column, and apply a precise claim to the resulting number. The distractor rows are placed there for a reason: they look almost right, but a row is from the wrong segment, the wrong region, the wrong year, or the wrong product line.

GMAT Focus Data Insights rewards reasoning over arithmetic, and table analysis is the purest expression of that philosophy. The integrated reasoning era introduced this style, and the Focus edition keeps it because it discriminates well. Two candidates with identical quant skills can finish in the same minute count and pick different answers because one treats the table as data and the other treats it as context. The candidate who wins is the one who reads the prompt first, sketches the column map, and only then touches the grid.

A useful framing: the table is a contract. Each column has a header, and the header governs every number beneath it. The header is the rule; the rows are the applications. If you forget the rule, you misread every row. The first ten seconds of any table analysis item should be spent reading every header, including any unit notation in parentheses, any footnote markers, and any segment labels. Most of the errors my students make on this question type start with a header they skimmed.

Why the prompt comes before the table

Candidates habitually read the table first because the table is bigger. That instinct is wrong. The prompt is the lens. Without the prompt, the table is just a rectangle. With the prompt, the table is a filter that turns a fifty-row grid into a two-row comparison. Read the prompt, underline the entities, and only then open the table to find them. The GMAT Focus does not penalise you for looking at the table; it penalises you for looking at the table without knowing what you are looking for.

The anatomy of a GMAT Focus table and spreadsheet prompt

Most table analysis prompts share a small set of structural features, and once you can name them, the whole item shrinks. The prompt typically names a business entity, a metric, a time period, and a comparison operator. The table then provides the data required to evaluate that comparison. The three statements each make a specific claim, and the candidates' job is to verify or refute each one against the grid.

For most candidates reading this for the first time, the entity is a company, a region, a product, or a demographic group. The metric is usually revenue, units sold, market share, growth rate, cost, or margin. The time period is often a year, sometimes a quarter, and occasionally a year-over-year change. The comparison operator is the verb that determines whether a statement is true: 'exceeds', 'is lower than', 'ranks third from the top', 'matches', 'differs by more than', or 'is closest to'. Marking that verb before opening the table is the single highest-leverage habit you can build on this question type.

The statements themselves follow a small number of patterns. Some are direct numeric claims: 'Region A's revenue in 2022 exceeded Region B's revenue in 2022.' Some are rank claims: 'Product X had the third-highest growth rate in the table.' Some are difference claims: 'The gap between Segment 1 and Segment 2 widened between 2021 and 2022.' A few are aggregate claims: 'The total of all regions in 2022 was greater than the total in 2021.' The mistake candidates make is treating each statement as a fresh puzzle. They are not. They are siblings that share most of the data. Once you have located the right rows for the first statement, the next two are usually one click away.

Headers, footnotes, and the contract of the grid

The contract of the grid is its header row, and footnotes are the exception clauses. A table that says 'Revenue ($ millions)' carries a unit annotation that you ignore at your peril. A table with a footnote marker on a single row means that row is governed by a separate rule, often because the number was restated, the segment was reclassified, or the figure is an estimate. If a statement concerns that row, the footnote controls. Most wrong answers in this question type come from treating a footnote-marked row as a normal row.

Sortable tables add one more wrinkle. The candidate can click a column header and re-sort the table ascending or descending. The exam never requires sorting to find the answer, but sorting often makes the answer easier. A rank claim like 'third-highest revenue' is faster to verify by sorting the revenue column descending and counting. A 'lowest cost' claim is faster with an ascending sort. Use the sort tool the way you would use a highlighter: as a way to bring a comparison to the surface, not as a substitute for reading.

The 40-second reading plan for the prompt

Speed on this question type is not about calculation speed. It is about reading speed, and the GMAT Focus gives you a finite time budget. If you spend two minutes on the prompt, you are already in danger. The plan I teach is a 40-second pass over the prompt that ends with three artefacts: a marked entity, a marked metric, and a marked comparison operator. Everything after that is verification.

Second 0 to 10: Read the prompt once, slowly, and name the entity in plain English. If the prompt says 'In the most recent fiscal year, the operating margin of the consumer segment exceeded that of the enterprise segment', the entity is the consumer segment versus the enterprise segment, the metric is operating margin, and the time is the most recent fiscal year.
Second 10 to 20: Locate the relevant columns in the table. Often they are obvious from the headers. If the table shows revenue, units, and margin, the answer is the margin column. If the table shows quarter-by-quarter revenue and the prompt asks about the most recent fiscal year, you may need a row sum across quarters before the comparison.
Second 20 to 30: Note the comparison operator. Exceeds, is lower than, ranks, differs by more than. This operator decides what your answer will look like: a strict greater-than, a rank position, or a magnitude difference.
Second 30 to 40: Skim each of the three statements and mark the first numeric claim in each. The first numeric claim is usually the most direct; the second and third are often the distractor statements designed to look like the first.

This 40-second pass buys you a problem that is already half-solved. You know what you are looking for, where to find it, and what shape the answer will take. The remaining 80 to 100 seconds are spent at the table, not the prompt.

Sortable versus fixed layouts: how the screen changes your method

The GMAT Focus presents tables in two formats. In a fixed layout, the table is a static grid and the candidate reads the numbers where they sit. In a sortable layout, the candidate can click any column header to re-sort the table. The two layouts are not the same question, and the optimal method differs. Treat the layout as a signal about the comparison the test-maker is asking for.

Fixed layouts favour direct comparison statements. The candidate looks up two rows, reads two numbers, and applies the operator. There is little benefit to re-ordering the rows because the comparison is positional. If the prompt names two regions and a year, the table will show those regions in some order, and you simply find them. In a fixed layout, your time is best spent on the headers, not the rows.

Sortable layouts favour rank and extremum statements. If the prompt asks which product had the highest growth rate, sorting the growth-rate column descending surfaces the answer immediately. If the prompt asks which region had the lowest cost, an ascending sort does the same. In a sortable layout, the table is interactive, and your method should be interactive too. A common error is to try to answer a rank question in a sortable table without sorting. The numbers are not organised for you, and forcing a rank from an unsorted grid is the surest way to pick the second-place row by mistake.

How the integrated reasoning ancestry still shows

Table analysis has its roots in the integrated reasoning table-analysis item, and the GMAT Focus keeps the older item's structure intact while changing the scoring weight and pacing. The three-statement format and the sortable-versus-fixed distinction both come from that older item. Candidates who trained on older prep material will recognise the format. Candidates who trained only on Focus mocks should practice both layouts explicitly, because the same stem can be presented either way, and the timing feels different.

Column signals that quietly decide your answer

Once you are at the table, your attention should fall on a small set of column signals. These are the features that, if you read them, make a statement true or false, and if you miss them, do the opposite. Six signals cover most of the items in circulation.

Unit annotations. A column marked in millions behaves very differently from a column marked in thousands. A difference of 50 in a thousands column is a rounding error. A difference of 50 in a millions column is material.
Footnote markers. Asterisks, daggers, and superscript letters indicate that the row or column is governed by an exception. Always read the footnote text. It often restates the figure or qualifies the segment.
Time stamps. A column for fiscal year 2022 is not the same as a column for calendar year 2022. A column for trailing twelve months is not the same as a column for the most recent quarter. The prompt often hinges on which time definition is in play.
Segment labels. A row labelled 'Consumer' may include or exclude a sub-segment called 'SMB'. A row labelled 'Enterprise' may include or exclude government. The labels are precise, and the prompt often exploits the precision.
Estimated versus actual markers. Some tables mark forecasts in italics or with brackets. A statement that says 'exceeded the forecast' is governed by the forecast column, not the actual column.
Sign conventions. A column labelled 'Change' can be positive or negative. A column labelled 'Margin' can be a percentage or a basis-point figure. The convention is not always obvious from the header, and the prompt is constructed to catch a candidate who assumes.

A worked example in column reading

Consider a prompt that says: 'In fiscal year 2022, the operating margin of the consumer segment exceeded the operating margin of the enterprise segment by more than 200 basis points.' A candidate who skims the table might read margin as a percentage, divide consumer margin by enterprise margin, and conclude the statement is true. The header says 'Operating margin (bps)'. The figure is in basis points, not percentage points. The candidate has just compared two numbers in the same unit, but the prompt was checking a 200-basis-point difference, and the table values are already in basis points, so the comparison still holds numerically. The error avoided is reading 'margin' as a percentage when the column is in basis points, a common error when the column header is short.

Now consider a second statement on the same prompt: 'In fiscal year 2022, the consumer segment had the third-highest revenue of all segments in the table.' A candidate who reads the revenue column and counts from the top might pick the wrong row because the column mixes fiscal and calendar year. The table has two revenue columns, one for fiscal 2022 and one for calendar 2022, and the prompt names fiscal. The rank claim is false because the candidate counted the calendar column. Reading the column header saves the question.

Distractor rows: where the GMAT Focus hides the wrong answer

Distractor rows are the rows that look right but answer a different question. They are not random noise. They are placed with intent, and the intent is to catch a candidate who knows the columns but not the rows. The most common distractor pattern is a row that matches every column except one. The candidate reads four out of five filters and lands on the wrong row, and the wrong row gives a number that makes a true statement look false or a false statement look true.

The way to neutralise distractor rows is to apply every filter the prompt names, in order, before you read a number. The filters are usually a segment, a time period, a region, and a product. If the prompt names two of these, you must satisfy both. If it names three, you must satisfy all three. A useful self-check: after you identify a row, glance back at the prompt and confirm that the row matches every filter. If even one filter is off, the row is a distractor.

Sortable tables make distractor rows easier to fall into. When the table is re-sorted by a different column, the distractor row can move next to the correct row, and a candidate who is reading by adjacency can confuse the two. The fix is to read the row label, not the row position. A row that says 'Consumer, SMB, 2022' is not the same as a row that says 'Consumer, Enterprise, 2022', even if they sit next to each other after a sort.

How statement patterns reveal the distractor

Statements come in clusters, and the cluster structure tells you where the distractor lives. A common cluster has one direct statement, one rank statement, and one aggregate statement. The direct statement usually involves a specific row. The rank statement involves ordering. The aggregate statement involves summing or averaging across rows. The distractor row is often placed to break exactly one of the three statements while leaving the other two clean. If two statements are true and one is false, the false one is the distractor. If two are false and one is true, the true one is the answer. Read the cluster as a whole, not statement by statement, and the pattern becomes visible.

Common pitfalls and how to avoid them

Most GMAT Focus table analysis errors fall into a small number of patterns, and each pattern has a corresponding fix. If you are making the same error twice in a row, you are not having bad luck. You are misreading the table in a specific way, and the fix is mechanical.

Reading the table before the prompt. The table is big, the prompt is small, and the brain reaches for the big object first. Reverse the order. Prompt, then table. The fix is a habit, not a tip.
Ignoring the unit annotation. A column in millions is not a column in thousands. A column in basis points is not a column in percentage points. Read the unit every time. The fix is to underline the unit the first time you read the header.
Skipping the footnote. Footnotes exist because the rule is different for one row. If a statement concerns that row, the footnote controls. The fix is to scan for footnote markers before reading any number.
Picking a row that matches most filters. A row that matches three of four filters is a distractor. The fourth filter is the one the test-maker wants you to miss. The fix is to apply every filter, in order, before reading the number.
Answering a rank claim without sorting. A sortable table is sortable for a reason. If the statement asks for a rank, sort the column. The fix is to recognise rank claims and respond with the sort tool.
Re-deriving the comparison from memory. After you read a number, hold it in working memory only as long as needed. The fix is to write the two numbers down, or to use the on-screen marker, before applying the operator.

A second tactical note on time

Most candidates reading this will benefit from budgeting around two and a half minutes per table analysis item. The GMAT Focus Data Insights has twenty items in forty-five minutes, and the table is one of the higher-cost items. If a table is taking you north of three minutes, you are reading it wrong, not slowly. Stop, re-read the prompt, and reset. Three minutes on a table analysis item is a strong signal that you have lost the column map and need to rebuild it.

Comparison with other Data Insights item families

Table analysis is one of five item families in the GMAT Focus Data Insights section, and the way it differs from the others matters for preparation. Each family has a different reading load, a different calculation load, and a different failure mode. The table below summarises the comparison.

Item family	Primary skill tested	Reading load	Calculation load	Typical failure mode
Tables and spreadsheets	Structured data lookup, column logic	High (headers and footnotes)	Low (one or two operations)	Misreading the column header or the row filter
Graphics interpretation	Reading a chart or graph and matching claims	Medium (axes and legend)	Low to medium	Misreading the axis scale or the chart type
Two-part analysis	Solving for two unknowns and selecting two answers	Medium	Medium	Forgetting that the two answers are linked, not independent
Multi-source reasoning	Integrating across tabs and discarding irrelevant tabs	Very high (three to four tabs)	Low	Reading a tab that does not apply to the question
Data sufficiency	Judging whether a statement is enough to answer	Low to medium	Varies	Treating the statements as full problems and solving them both

Reading the table, the picture is clear. Table analysis has the highest reading load per item and the lowest calculation load per item. If you are a fast calculator and a slow reader, the table will hurt you. If you are a slow calculator and a fast reader, the table will reward you. Preparation should be aimed at the reading load: training your eye to find the right column and the right row in the shortest possible time.

A practice protocol for the next ten sessions

Reading the table is a skill, and skills need a protocol. The protocol I give my own students is a ten-session loop that takes about thirty to forty minutes per session. The aim is not to grind through hundreds of items. It is to install the 40-second reading plan, the column-signal scan, and the distractor-row filter as automatic behaviour.

Sessions 1 to 2: Practice the 40-second reading plan on ten items per session without timing the table. The goal is accuracy of column identification, not speed.
Sessions 3 to 4: Add a three-minute total budget per item. If you exceed three minutes, stop and re-read the prompt. Track which filter you missed.
Sessions 5 to 6: Practice the sortable layout specifically. On every item, decide in five seconds whether the layout is sortable. If yes, sort before you answer any rank claim.
Sessions 7 to 8: Practice the fixed layout. On every item, identify the unit annotation and the footnote marker before reading any number. Write the unit on your scrap paper.
Sessions 9 to 10: Mixed items under exam timing. Twenty items in forty-five minutes, with a target of two and a half minutes per table and the remainder for the other four families.

Tracking the right metric

Most candidates reading this will measure progress by accuracy. Accuracy is the wrong leading indicator on table analysis. The right leading indicator is the number of times per item that you re-read the prompt. If you are re-reading the prompt more than once, you are reading the table without a filter. If you read the prompt once and apply every filter, you do not need to re-read. Track the re-read count, not the accuracy, and the accuracy will follow.

Putting it all together on a real item

Read the table from start to finish, and you will see how the pieces fit. The prompt names a segment, a metric, and a comparison. The 40-second pass marks the entity, the metric, and the operator. The column-signal scan locates the unit annotation and the footnote marker. The distractor-row filter applies every prompt filter in order. The sortable-versus-fixed decision decides whether to sort. The cluster structure of the three statements tells you which one is the distractor. Two and a half minutes later, you have answered an item that most candidates answer by guessing on the third statement.

This is the work of table analysis on the GMAT Focus. It is not glamorous, and it does not require clever arithmetic. It requires the discipline to read the prompt first, the patience to scan every header, the eye to apply every row filter, and the judgement to use the sort tool when the layout is sortable. Candidates who build that discipline separate themselves from candidates who treat the table as decoration. The 200 to 300 points of Data Insights performance that separates a strong score from a middling one often comes from this exact question type, and a candidate who has installed the 40-second reading plan is operating with a different floor than the candidate who has not.

For most candidates, the next step is not more items. It is a structured diagnostic. TestPrep Europe's diagnostic assessment is a natural starting point for candidates building a sharper preparation plan around the GMAT Focus tables and spreadsheets question type, with a particular focus on column reading, distractor row filtering, and the sortable layout.

Frequently asked questions

How long should I spend on a GMAT Focus table analysis item?

Target two and a half minutes per item, including the prompt read. If you pass three minutes, you have lost the column map and should re-read the prompt rather than continue searching the table. The reading load is high but the calculation load is low, so a clean first pass usually beats a long second pass.

What is the difference between a sortable and a fixed table?

A fixed table keeps its original row order, and you read numbers in place. A sortable table lets you click a column header to re-sort ascending or descending. Sortable tables are typically used for rank and extremum claims, where sorting surfaces the answer faster. Use the sort tool as soon as the statement asks for a rank or a highest or lowest value.

Do I need to perform any heavy arithmetic on table analysis?

No. Most statements require at most one subtraction, one division, or a quick percentage. The GMAT Focus rewards reasoning over arithmetic, and table analysis is the purest example. The work is in the column reading and the row filter, not in the calculation.

What is the most common error on this question type?

Reading the table before the prompt, then applying most but not all of the row filters. The result is a distractor row that matches three of four filters and gives a number that flips the statement. Apply every filter the prompt names, in order, before reading the number.

How does table analysis differ from graphics interpretation?

Table analysis presents a structured grid of numbers with row and column headers, and the work is column reading and row filtering. Graphics interpretation presents a chart, graph, or visual, and the work is reading axes, legends, and chart type. Both have a low calculation load, but tables reward careful reading of headers, while graphics reward careful reading of scales and legends.

GMAT Focus tables and spreadsheets: 6 column signals that decide your answer