Timing & strategy: 40 questions, 60 minutes, no panic

Academic Reading gives you 60 minutes for three passages and 40 questions, and - crucially - no extra time to transfer answers (on the paper test, the answers you write are the answer sheet). Plenty of capable readers under-score purely because they run out of time on Passage 3. A plan fixes that, and the plan matters more than your raw reading speed.

The marks-per-minute mindset

There are 40 marks and 60 minutes, so each question is worth about a minute and a half of your time - no more. Every mark counts the same: a brutal Passage 3 inference question is worth exactly as much as a one-word Passage 1 detail. It follows that the worst possible trade is to spend three minutes rescuing one hard mark while three easy marks slip away unread. Hold that idea and most timing decisions make themselves.

A workable pacing plan

• ~20 minutes per passage, but bank time early. Passages get harder, so aim for 17-18 min on Passage 1, ~20 on Passage 2, leaving 22-23 min for Passage 3.
• Each mark is worth the same. Never let two stubborn questions eat the time of ten gettable ones.
• Leave 2-3 minutes at the end to make sure every line has an answer.

A simple checkpoint table to glance at:

The method, step by step

1. Skim the passage (~2 min) and label paragraphs before reading questions (see Skimming & scanning).
2. Do the question types in the order that suits you, not necessarily the printed order. Matching headings and TFNG reward a fresh memory of the whole text; detail and completion questions can be scanned at any time.
3. Set a hard limit per question: ~1.5 minutes. If you're stuck, guess, flag it, move on - there's no negative marking, so a blank is a wasted lottery ticket.
4. Watch the clock at fixed checkpoints, not constantly: glance at 18 and 38 minutes. If you're behind, speed up immediately rather than at the end.
5. Final sweep: fill every blank with your best guess. 40 answers attempted, always.

Worked example 1 - triage in action

You're 18 minutes in and still have two TFNG items on Passage 1 plus all of Passage 2. Decision:

• A TFNG you're "60% sure" on → mark your best guess, flag it, move to Passage 2. Coming back later with fresh eyes is cheaper than grinding now.
• One question needs a number you cannot find after 90 seconds → guess and go. Spending three minutes to maybe gain one mark risks losing several easy Passage 3 marks.

The candidate who finishes all three passages and guesses six hard ones almost always beats the one who "perfected" Passage 1 and never reached Passage 3.

Worked example 2 - sequencing question types within a passage

Passage 2 has, in printed order: an MCQ set, then matching headings, then a summary completion. A strong order of attack is:

1. Matching headings first, while your skim of the whole passage is fresh - it tests the gist you've just built.
2. Summary completion next, scanning for the gaps (fast, anchored on the passage order).
3. MCQ last, since each question needs slow close reading of specific lines anyway.

You answered the same questions, but you did the memory-dependent task while memory was hot and saved the slow task for when you'd already located most of the text. Doing them in printed order would have wasted your fresh overview.

Common mistakes

• Reading the whole passage first. That alone can burn 8-10 minutes. Skim, then let the questions drive your reading.
• Sunk-cost questions. Refusing to leave a hard item because you've "already spent two minutes". Cut your losses and flag it.
• No final guesses. Blanks score zero; a guess might score, especially on TFNG and MCQ. Never hand in an incomplete sheet.
• Checking the clock every 30 seconds. Constant clock-watching breaks concentration; trust your two fixed checkpoints instead.
• Saving transfer for the end (paper test). There is no transfer time - write answers straight onto the sheet as you go.

Try it

At /ielts/reading, do a full timed set with a visible timer. Note the clock when you start each passage and compare it to the checkpoint table above. Afterwards, check: did any single question take over two minutes? That's the leak to plug next time.