Read the Instructions
Exams are stressful occasions and stress makes people behave in odd ways. This includes failing to do obvious things like reading the instructions at the top of the exam paper and following those instructions. A surprisingly large proportion of people have at one time or another failed to read the instructions on an exam paper or have failed to follow them and thus severely reduced their chances of success. Don't be one of those people! The very first thing you should do is to read every word on the exam paper carefully and be clear what is being asked of you - e.g., how many questions to answer, and from which sections of the paper, etc. Don't assume that you know the instructions for the paper already - they might have changed this year, or you might have been misinformed.
Look at Both Sides of the Sheet
Another very obvious point — look at both sides of the exam paper (or both sides of each sheet if there is more than one sheet). Make absolutely sure that you have looked at the whole paper and have not missed a couple of extra questions or a whole new section over the page.
It is essential that you divide your time evenly between each of your essays during the exam. You will seriously damage your chances of success if one of your essays is exceedingly short, scrappy, or non-existent. You should divide the total time available by the total number of essays you need to write. So, for example, if you have 3 hours to write 4 essays then you have 45 minutes per essay. Get a scrap of paper and write on it the time you need to start each new essay: e.g. '3.00 Essay 1, 3:45 Essay 2, 4:30 Essay 3, 5:15 Essay 4'. Stick to these times rigidly - even taking 5 or 10 minutes too long for each of the first 3 essays would leave you with nowhere near enough time for your final essay. Always be aware of how long you have had and how much time you have left for each essay.
Allow time for planning and reading through
Before writing an essay in the exam, spend two or three minutes preparing a rough plan. If you have memorised a list of main themes for a topic, or some dates and examples, or quotations, then now is the time to scribble them all down on paper. Have a minute or two doing this, brainstorming, thinking of all the important points you need to mention in your answer (whether you have deliberately memorised any or not). Then draw up a very simple plan with just a word or two for each main paragraph. Make sure that your planned essay is going to answer the question that has been set, and be clear in your mind what your answer is and what sort of conclusion you are working towards. You should also include a rudimentary map at the start of your essay to help the examiners find their way through your answer more easily. The lack of time in an exam situation means that the map will be less detailed than it would be for a supervision essay, but it should still be there. Essays in which examiners get lost and disoriented get lower marks than easily navigable essays.
Also allow a few minutes at the end of each essay for reading it through to eliminate slips and blunders (spelling, grammar, missing words etc.). The fact that you are writing under pressure of time makes it more likely that you will make mistakes and therefore more important to go back and check.
Answer the question
Memorising lists of major points for each topic is probably a good idea. However, it is a very bad idea simply to find the title that roughly corresponds to your topic, write it at the top of your page, and then disgorge a pre-prepared planned essay. Even though spotting likely topics is pretty easy, it is impossible to predict what actual question on the topic will come up. It is not that likely that a `plan' you memorise for a topic will fit exactly the question that comes up. DO NOT JUST REPRODUCE A PREPARED ESSAY IF IT DOES NOT FIT THE QUESTION. Always answer the question. If your prepared list of themes and points on a topic don't really fit the question, there is no need to despair or panic, just spend a few minutes thinking about how to tailor the knowledge and understanding you have, and how to rearrange your points (which ones to discard, what new material to bring in etc.) so as to answer the actual question that has been set. It is better to write an essay that really answers the question that might be, say, a little short on examples, than to write a well-prepared essay with lots of examples that answers a totally different question.
Information, not waffle
Some exam questions are very general, abstract, or purely philosophical and, do not explicitly ask for facts, dates, quotations, or references to primary and secondary material, e.g. 'Would a world based on chance be incompatible with a purposeful God?' Do not be fooled! Do not give a vague, general, waffly answer. The examiners expect you to answer the question with reference to the knowledge you have acquired from your lectures, reading, and supervisions. They expect you to show that you are familiar with the most important arguments on the topic, who made them, and in which books. They also expect you to be able to provide examples for any general points you make.
Some people think that referring to a primary or secondary text is a bad idea, since it shows that you are 'just' repeating what other people have said. But that is what you are required to do in an exam – that is precisely how you indicate that you have learned things during your course and have taken them on board. You will always get credit for referring to a primary or secondary text by name and showing that you know how one thinker differs from another in the way he approaches a problem. The only way that referring to primary or secondary texts could be anything other than a source of credit would be if it was all you did.
Analysis as well as Information
As well as showing that you have learned the relevant facts and arguments, you are required in exam essays – as in supervisions essays – to provide your own analysis. The essay should always be driven by your own argument; you also need to explain and analyse others' arguments.