Google Forms is a good quiz tool and a poor exam tool. For a feedback form, a worksheet check or a low-stakes recall quiz it is free, already in your school account, and completely adequate - use it and stop reading. For a competitive-exam mock it gives out at the first thing that paper needs. There is no native timer, no negative marking without a script, no sectional timing, no question palette or mark-for-review, no way to lock a section once it is done, and the results come back as survey responses rather than an item analysis.
None of that is a fault. Forms was built to collect answers from people, and it does that job well. Quizzes mode holds an answer key, marks a right answer right, totals the score and shows it the moment a student presses submit. For a chapter check on a Tuesday that is the entire job, and moving off Forms to do it would be a waste of an afternoon. The question is not whether Forms is good. It is whether the thing you are building is a quiz or an exam.
What Google Forms Is Genuinely Right For
Keep Forms for anything where the clock is not part of what you are measuring. The moment the output you want is a list of responses rather than a rank list, Forms is the correct tool and anything heavier is overhead.
- Feedback forms, consent forms, registrations, and anything whose output is a spreadsheet of replies.
- Homework and worksheet checks, where you care that the work was done and roughly how it went.
- Recall quizzes - vocabulary, formulae, dates - run open-book or untimed on purpose.
- A quick diagnostic before you start a chapter, where you want a show of hands rather than a merit order.
- Anything a student will fill in on a phone, in a hurry, from a link in a parents' group.
Where an Exam Paper Breaks It
A competitive-exam mock is not a longer quiz. It is a rehearsal, and what it rehearses is pressure under rules. Strip the rules out and the student practises something that is not the exam. There are six places a form builder gives way, roughly in the order a paper setter meets them.
- No clock. Nothing counts down, nothing stops the student at the end, and nothing records how long they actually took.
- No negative marking. A wrong answer scores zero, never minus one, unless somebody writes code to re-score the sheet afterwards.
- No sectional timing. One form is one undivided run; you cannot hand Section A its own forty-five minutes.
- No question palette. A student cannot see at a glance which questions are answered, which were opened and left, and which are flagged.
- No mark for review. There is no way to flag a question, move on, and come back to it, which is how a long paper is actually sat.
- No section lock. Nothing stops a candidate scrolling back into a part of the paper the real exam has already closed.
The Timer Is the First Wall, and the Add-On Is the Second
Forms has no countdown of its own. The usual fix is a third-party add-on, and the trade is worth naming out loud: that timer is not Google's, it belongs to whoever wrote the add-on, and authorising it means granting a third party access to the form and the responses sitting in it. For a school running a mock on minors' data, that is a decision somebody should make deliberately rather than a checkbox somebody ticks. Adding a timer to Google Forms goes through the workarounds and where each one leaks.
A clock built for an exam has to do more than count. On MockSetu the countdown runs in a background worker, so a student who switches tabs or locks the phone does not return to a timer that quietly stopped. A warning fires with five minutes left. At zero the paper submits itself, banking the time spent on the question that happened to be open, so nothing is lost to the instant the clock ran out. If the device dies mid-paper there is a five-minute window on the same device to get back in and land on the question they were on - with whatever time is left, because the clock keeps running while they are away. Stay away longer than that and the sitting is sealed and filed as a normal attempt. How to create a timed online test with auto-submit sets that up end to end.
Negative Marking Is a Script, Not a Setting
Forms adds points up. It does not take them away. The workarounds are to export the responses and re-add them somewhere else, or to write a script that re-scores on submit. Both work. Both mean the number the student sees when they press submit is not the number that counts, which removes most of the reason for running a mock in the first place - a candidate learns the price of a guess by paying it, immediately, on screen. Negative marking workarounds in Google Forms covers what each route costs you.
A tool built for exams treats marking as three numbers on a question: what a right answer is worth, what a wrong one costs, and what leaving it blank costs. Four presets cover the common schemes and you type anything else yourself. Because a question can carry its own numbers, the rule resolves question first, then section, then the exam default - so one paper can carry two completely different schemes. SSC MTS is the standard case: 90 questions for 270 marks across two sessions of 45 minutes each, where Session I is qualifying only with no negative marking and Session II counts towards merit and deducts one mark. A single scheme for the whole form cannot reproduce that paper. It reproduces something that looks like it and scores wrongly.
A mock that scores a wrong answer at zero when the real paper charges a mark is not an easier mock. It is a different exam, and the habit it teaches costs marks on the day.
The Screen the Student Sits Is Part of the Paper
Mark for review and the question palette are not conveniences. They are how a candidate manages a long paper against a short clock: flag the ones you half-know, clear everything else, come back with whatever is left. A paper practised without them trains a strategy the real console will not support. On an exam screen the palette colours every question - green for answered, purple for opened and left, red for flagged, plain for never touched - and a Mark for Review button sits at the top of every question, level with the question-type chip. Section lock matters for the same reason. When the real exam closes a part, the mock has to close it; when two subjects share one pooled clock, the mock has to pool them. A form that scrolls top to bottom in one column cannot rehearse any of it.
A Response Sheet Is Not an Item Analysis
This limit bites after the paper, when you are deciding what to teach next. A form's results view is built for a survey: here is the spread of answers to each question, here is the average score. Useful. It is not the same as knowing which question in your paper is quietly broken.
A creator analytics page answers different questions. Per question it gives accuracy and average time, averaged across everyone who attempted that question. It lists the most skipped questions, the ones most often marked for review, and - open this one first - the questions most people got wrong, next to the wrong option that was picked most. A whole batch landing on the same wrong option is a misconception with a name, and you can teach it on Monday. Accuracy also breaks down per section, which tells you whether the problem was the chapter or the paper.
Building the Paper Once You Have Decided
The build order is short and it matters, because timing and marking both hang off sections.
- Create the exam and its sections first, each with its own minutes, before a single question goes in.
- Get the questions in. If the paper already exists as a PDF, run the published extraction prompt from the JSON upload guide in whatever AI you already use and upload the output - knowing that numeric, TITA and match-the-column questions arrive as placeholders for you to finish by hand.
- Set the exam default marking scheme, then override only the sections that genuinely differ from it.
- Write the scheme into the instructions in words, section by section. Publishing checks the timing sentence and the section and question counts in that text against the paper, but nothing proves the marking numbers in your wording right - change a penalty later and you re-read the text yourself.
- Preview it yourself and sit a few questions. A creator preview records nothing - no attempt, no score, no row in the analytics.
- Publish. A paper with an empty section, a blank question, an option list shorter than two, or a missing answer key is refused until you fix it.
What This Tool Cannot Do Either
Being honest about Forms is worth nothing if the replacement oversells itself. So: there is no proctoring here of any kind - no webcam, no lockdown browser, no tab-switch detection. Published papers are public, which means anyone with the link may attempt them and there is no paid or private delivery to one batch. There is no CSV or Excel export of results and no per-student report card. No certificates. No email, SMS or WhatsApp alert telling a batch a test has gone live, so you still send the link yourself. There is no question shuffling and no cap on attempts - two controls a form builder restricted to signed-in respondents handles more naturally than this does, and if those are the things you need most, that is a genuine argument for staying where you are.
So the test is simple. Are the clock, the penalty and the order of the paper part of what you are measuring? If not, Forms is right, free and already open in another tab. If they are, you want sections with their own clocks, three marking numbers per question, a palette the student will recognise from the real console, and an analytics page that names the question your class got wrong. Everything a creator can build sits behind one free account that takes no card, and a published paper gets a listing in the public library of mock tests where students can actually find it.