GMAT® Word Problems: Half Math, Half English, and the Setup That Prevents Translation Mistakes
If your gut instinct on a GMAT® word problem is to read the question and go straight to equations, that makes sense.
Most of us learned math that way. Read the problem. Set up the equation. Solve.
On the GMAT®, that path costs more points than any other single habit on the Quantitative section.
The reason is that most word problem mistakes happen before the math even starts. The arithmetic is usually clean. The algebra is usually doable. What goes wrong is the translation — the moment between reading the prose and writing the equation, when the brain is asked to do two things at once and a key detail slips through.
The fix is a small intermediate step. We call it "half math, half English." It is one of the highest-leverage habits we cover in the podcast, and the three problems in Episode 43 of Real GMAT® Problems, our podcast series show why.
What Half Math, Half English Looks Like
The idea is to write down what the problem is telling you and asking you — using a mix of math notation and plain English — before you try to make full equations.
For an average revenue problem, that might look like:
- 10-day average = $400
- 6 days average = $360
- last 4 days average = ?
That is not fully math. It is not fully English. It captures the structure of the problem without forcing the brain to do the translation in one step.
For a tiered pricing problem, it might look like:
- 1 to 5 cases = $60 each
- 6 to 20 cases = $50 each
- more than 20 cases = $45 each
- 3 cases + 11 cases + 30 cases = total charge ?
Again — partial math, partial English. The structure is on the page. The full equations come next.
The benefit is bandwidth. Short-term memory can hold about seven things at once, and word problems often have more than that. Forcing the brain to read, translate, and write equations all at the same time is what causes the careless mistakes. Splitting the work into stages — read, capture, then equate — frees up bandwidth at each step.
We use this on the snack shop average revenue problem at the start of the episode, and the same structure carries through to the harder questions.
The Average Formula Bridges Most Word Problems
The average formula — sum / count = average — comes up in nearly every word problem episode we record.
The reason is that the formula works in both directions. Given any two of the three quantities, you can solve for the third.
For most word problems, the missing piece is the sum. You may be given an average and a count, and the next step in the chain is converting that into a fixed total. That total is what unlocks the rest of the problem.
This shows up directly in the warm-up and the hard problem from this episode. Both rely on the same move: average × count = sum.
The Trap on Tiered Pricing
The second problem in this episode catches more people than the warm-up — by a lot.
The setup describes pricing brackets: one price for small orders, a lower price for medium orders, an even lower price for large orders. Three separate orders are placed. The question asks for the total charge.
The trap is reading the brackets as cumulative — as if a 20-case order pays the highest rate for the first 5 cases, then steps down for the next 15. That is how some real-world pricing structures work. It is not how this problem is set up.
The pricing here is siloed. Each order, taken as a whole, gets one rate based on its total size. A 30-case order is 30 × $45. No splitting, no stepping down.
Most of the wrong answers on this problem cluster on the option that comes from the cumulative interpretation. The math is being done correctly. The setup is wrong. We walk through the full setup on the beverage distributor problem page.
The lesson generalizes. When a problem describes a structure (pricing, scoring, rates, anything that varies by quantity), check whether the structure is siloed or cumulative before you start computing. The half math, half English step is where that check naturally happens.
Rows and Columns Scale With Complexity
The hardest problem in the episode is the library books question — 30 students with different borrowing counts, a class average, and a question asking for the maximum any single student could have borrowed.
This problem has more data points than short-term memory can hold comfortably. The fix is a table. Two columns — one for the number of students, one for the books they borrowed in total. One row per group. Question marks where the value is unknown.
The table does not solve the problem. It keeps the data visible so you can see which pieces are still missing and which operations still need to happen. On a problem with this many parts, that visibility is what prevents you from finishing the math and then realizing you solved for the wrong quantity.
If you have not been using tables on multi-data-point problems, this is a good place to start. The habit is easy to build on easier problems and pays back on the harder ones.
What to Take Away
Three habits to carry into any word problem:
-
Write what is given and what is asked in half math, half English before writing any equations. Each line of the problem becomes one line on your scratch paper.
-
Use the average formula (sum / count = average) as the bridge whenever the problem gives you an average. Convert it to a fixed total so the rest of the problem has something to work with.
-
Use rows and columns for any problem with three or more categories of data. The structure prevents the careless mistakes that come from losing track of which number means what.
Most wrong answers on word problems are translation errors, not arithmetic errors. The three habits above are the highest-leverage protection against that pattern. None of them is fast on the first try. All of them get fast with reps.
FAQ
What is the "half math, half English" approach?
A structured intermediate step between reading a word problem and writing equations. You capture each piece of given information and the question itself using a mix of math notation and plain English — for example, "10-day average = $400" or "max books single = ?" The point is to get the structure on the page before doing any computation.
Why is the half math, half English step worth the extra time?
It frees up short-term memory. Word problems often have more data than working memory can hold at once. Splitting the read-then-translate process into two stages — capture first, equate second — reduces the cognitive load at each step and prevents the careless errors that cost more time later in double-checking.
What is the difference between siloed and tiered pricing on a GMAT® word problem?
In siloed pricing, an order's entire price is based on the total size of that order — a 30-case order is 30 × the single rate for that bracket. In tiered or cumulative pricing, an order is split across brackets — the first chunk pays the highest rate, the next chunk pays a lower rate, and so on. GMAT® problems usually describe siloed pricing. Read the wording carefully before assuming one or the other.
When should I use a table on a word problem?
Any time the problem gives you data across three or more categories — for example, students grouped by how many books they borrowed, or orders grouped by size. A two-column table with categories down one side and the relevant quantity across is usually enough. Rows and columns prevent the data-tracking errors that show up on long multi-step problems.
Most of my wrong answers are translation errors, not math errors. Is that normal?
For most students working on the GMAT® Quantitative section, yes. Translation errors are the more common cause of missed word problems than arithmetic mistakes — particularly on questions you would describe as "ones I should have gotten right." Building a repeatable setup process — half math, half English plus tables for complex problems — is the most effective fix.
How does the GMAT® scoring algorithm treat missed easy questions?
The GMAT® rewards consistent execution. Missing questions that match your demonstrated ability level is more damaging to the score than missing harder questions where the conceptual gap is the limiting factor. That asymmetry is why we focus so heavily on setup habits — they protect against the missed easies that hurt the score the most.
Want to Learn Even More?
Listen to the full episode on the platform of your choice:
- Episode 43 page on the blog (with embedded player)
- Spotify
- Apple Podcasts
- YouTube
The previous episode in the series, Episode 42 on statistics, covers the average formula and median definitions that the hardest word problem in this episode relies on.
The three worked solutions from Episode 43:
-
Problem 1 (warm-up): In a Small Snack Shop the Average Revenue Was $400 per Day. Half math, half English plus the average formula in both directions.
-
Problem 2 (medium): A Beverage Distributor Charges $60 per Case for 1 to 5 Cases. Siloed vs. cumulative pricing — the most common translation trap on this question type.
-
Problem 3 (hard): In a Class of 30 Students, 2 Did Not Borrow Any Books. A rows-and-columns table plus the average formula plus the optimization logic from the previous episode, all in one question.