The habit that got you through JEE will cost you here
Every Indian entrance exam you have ever sat rewards the same reflex: sweep the paper, bank the easy marks, flag the hard ones, come back with the time you saved. It is good advice. It is drilled into you for two years. And in NATA Part B it is worth nothing, because there is nothing to come back to.
One question, then the next, then the next
Part B presents questions one after another under a 108-second budget each. The Test Center Manual describes the flow this way, and there is no evidence in any official document of a review screen, a question palette, or a mechanism for returning to something you passed over.
Think about what that removes. No sweep. No triage. No flagging. No back button. The question in front of you is the only question that exists, and when it goes, it is gone.
This is not a small change of tactics — it inverts the strategy. In a linear paper, the cost of a hard question is time, which you can manage. In a one-way paper, the cost of a hard question is the question. You either answer it now or you never answer it.
What adaptive actually means for you
The other half is that the test is adaptive. The President's foreword to the 2026 brochure dates this to NATA 2025 and confirms 2026 continues it.
Be careful about what that does and does not imply. An adaptive engine adjusts what it serves you based on how you are doing. In practice this means two things worth internalising. First, the paper is not the same for everyone, so comparing your experience to a friend's afterwards tells you very little. Second, a run of hard questions is not bad luck — it may well mean you are doing well. Candidates who read difficulty as failure and panic mid-test are misreading the one signal the engine gives them.
What the Council has not published is how the adaptation works, how questions are weighted, or whether wrong answers are penalised. Anyone describing the algorithm to you is guessing.
The part nobody can tell you
Two facts you would want before choosing a strategy simply do not exist in the public record.
Marks per question are unpublished. 120 marks across 50 questions averages 2.4, which is consistent with variable weighting under an adaptive engine — but that is arithmetic, not policy. Whether wrong answers carry a penalty is also unpublished. The brochure is silent. The near-universal claim that NATA has no negative marking traces only to coaching sites; we could not source it to COA at all.
That second one has teeth, because it is exactly the fact a guessing policy depends on. If wrong answers are free, you should answer everything. If they are not, blind guessing is destructive. You do not know which world you are in — so do not build a strategy that only works in one of them. Answer what you can reason toward, including partial elimination. Do not fire blind at a question you have not engaged with just because someone on the internet told you it was free.
The rules behind this
Sourced to the official brochure rather than restated here, so there is one place to correct when the Council revises it.
Part B allows 108 seconds per question, presented one after another, on an adaptive engine.
90 minutes across 50 questions. The adaptive structure dates to NATA 2025 per the President's foreword in V2.0, which states that NATA 2026 continues it.
Source · verified 2026-07-16
Part B questions appear one after another at 108 seconds each. There is no evidence of a review screen.
The skip-flag-and-return habit that works in JEE does not transfer. Budget the 108 seconds and commit.
Source · verified 2026-07-16
Whether Part B carries negative marking is not stated in any official document.
The widely-repeated "no negative marking" claim appears only on coaching sites. We could not source it to COA.
Read this carefully: This one matters for strategy — whether to guess depends on it. Since it is unpublished, do not build a guessing policy on the assumption that wrong answers are free.
The marks awarded per Part B question are not published anywhere in the official brochure.
120 marks across 50 questions averages 2.4, which would be consistent with variable weighting under an adaptive engine — but that is arithmetic, not policy. Coaching sites uniformly assert 1-3 marks per question.
Read this carefully: The brochure does not state a per-question mark. Any specific mark value you see quoted is third-party inference presented as fact.
What almost everyone believes
“I will do a first pass for the easy ones, flag the hard ones, and come back at the end.”
There is nothing to come back to. Questions arrive one at a time and there is no evidence of a review screen.
This is the single most transferable-looking habit from JEE and the most damaging one here. A candidate running a triage strategy on a one-way test does not merely fail to save time — they answer hastily on the assumption of a second look that never comes, and lose marks on questions they could have got. Practise committing, because the exam will make you commit whether you have practised or not.
Depending on how long you have
Foundation
Understand the skill. Months out, or starting from zero.
Start building the commit habit early, because it is a habit and not a technique. When you practise any MCQ set, forbid yourself from skipping. Answer in order, move on, do not look back. It will feel worse than your usual method for a few weeks. That is the point — you are unlearning a reflex that took two years to install.
Drill
The practice protocol. What to repeat, how often, how to score it.
Every Part B practice session runs one-way: no skipping, no returning, hard timer, auto-advance. If your practice tool lets you navigate backwards, it is training the wrong exam. Track how often you would have gone back — that number falling is the actual measure of progress here, more than your score.
Exam-Day
What to actually do under the constraint — 108 seconds, no instruments, one pass.
Decide before you sit down that you will not fight a question. When one is beyond you, reason as far as you can, commit, and let it go — the next one is already the only one that matters. Do not read a hard run as a sign you are failing; under an adaptive engine it may mean the opposite. And do not spend the exam theorising about the algorithm.
Try it
Fifteen minutes. The aim is to feel the discomfort now rather than on 7 August.
- 01Take any set of 20 aptitude or reasoning questions you have not seen.
- 02Set a timer for 108 seconds per question — 36 minutes total — and answer strictly in order.
- 03Rule: once you move on, you may not return. Cover previous questions with paper if you have to.
- 04Note every question where you felt the pull to skip. That pull is the reflex you are unlearning.
- 05Score it, then ask a harder question: on how many did you commit too fast because you assumed a second pass? That number is your real starting point.
The short version
Part B is one-way and adaptive. You cannot skip, flag or return; a hard run may mean you are doing well rather than badly; and the two facts you would most want before choosing a guessing policy — marks per question and negative marking — are not published at all. Train the commit habit, and do not bet on the internet's assurance that wrong answers are free.
Next: what 108 seconds actually buys you, and how to spend it.
Questions people actually ask
- Can I go back to a previous question in NATA Part B?
- There is no evidence that you can. The Test Center Manual describes questions appearing one after another under a 108-second budget, and no official document mentions a review screen or question palette. Prepare on the assumption that you cannot return, because that is what the record supports.
- Does NATA have negative marking?
- The official brochure does not say. It is silent on both negative marking and marks per question. The widely-repeated claim that there is no negative marking appears only on coaching sites and cannot be sourced to the Council of Architecture. Do not build a guessing strategy on it.
- What does it mean if my NATA questions feel very hard?
- Possibly that you are doing well. Part B is adaptive, so the engine adjusts what it serves based on your performance. A run of difficult questions is not automatically a bad sign, and candidates who read difficulty as failure and panic are misreading it. The Council has not published how the adaptation works.
- Is NATA Part B the same paper for everyone?
- No. Because the test is adaptive, what you are served depends on how you are performing. Comparing your experience with a friend's afterwards tells you very little about how either of you did.