IFoA SA3 common mistakes
IFoA SA3 · Understand the traps and how to avoid them.
Spending the first 20-30 minutes reading the whole paper and planning before writing anything, then running out of time to finish the final question.
Candidates treat SA3 as a knowledge test where thorough planning earns marks, but marks come only from written points. The paper is long and every question carries substantial marks, so planning time is dead time.
What to do instead
Cap reading and planning at about 10 minutes total. Start writing the first question within 10 minutes of the exam starting, and allocate time strictly pro rata to marks (roughly 1.8 minutes per mark). Leave a two-minute buffer for checking.
Writing long paragraphs of prose that explain a concept correctly but never state the specific action, item, or conclusion the question asked for.
Candidates are used to essay-style writing and believe depth signals understanding. Markers award marks per distinct valid point, so a paragraph containing one point earns one mark regardless of length.
What to do instead
Write in bullet points, one idea per bullet, with the point in the first few words. Before moving on, re-read the command word ('describe', 'explain', 'recommend') and check each bullet directly answers it.
Producing fewer points than the marks available, for example writing six bullets for a 10-mark 'describe' question and stopping.
Candidates write until they feel they have covered the topic rather than until they have generated enough distinct points. They underestimate how many separate acceptable points a 10-mark question needs.
What to do instead
Use the rule of thumb of at least one distinct point per mark, plus two or three spare. Number your bullets so you can count them against the marks before moving on.
Answering a calculation question with a final number and no workings, or with workings that use unlabelled symbols and unexplained assumptions.
Candidates focus on getting the right answer and assume the marker will follow the arithmetic. In practice method marks are awarded for visible steps, and a wrong final number with clear workings still earns most of the marks.
What to do instead
Show every step, define each symbol, state assumptions (for example the distribution used or the independence assumption), and give the final answer with units on its own line. Do this even when the answer feels obvious.
Treating reserving questions as pure technique and omitting the business context: regulation, capital, reinsurance recoveries, expenses, discounting, and communication to the board.
Candidates revise the standard reserving methods heavily and then answer every reserving question with a method comparison. SA3 questions usually ask for judgement and communication, not just the mechanics.
What to do instead
For every reserving scenario, run a checklist: method choice and why, data quality, assumptions, uncertainty and range, reinsurance, expenses, discounting, regulatory and capital impact, and how you would explain it to a non-technical audience.
Ignoring the scenario details in the question stem (the company's size, portfolio mix, distribution channel, regulatory regime, or stated risk appetite) and giving generic textbook answers.
Under time pressure candidates pattern-match to a familiar topic and reproduce memorised lists. SA3 questions are deliberately written so that the specific facts change which points are valid.
What to do instead
Underline or note the three or four scenario facts before answering, and tie at least half your bullets explicitly to those facts (for example 'because the book is 70% motor, the reserve risk is short-tailed, so...').
Writing a balanced list of pros and cons when the question asks for a recommendation, and never actually recommending one option.
Candidates learn to 'consider both sides' and believe balance demonstrates judgement. Questions asking for a recommendation award marks for a clear decision plus justification.
What to do instead
If the command word is 'recommend', 'advise', or 'propose', state the chosen option in the first sentence, then justify it against the scenario's stated objectives, then note briefly why the alternatives were rejected.
Spending disproportionate time on the first question, which often covers familiar pricing or reserving material, leaving too little time for later questions on capital, reinsurance, or regulation.
Candidates start with what they know best and lose track of time while writing many points on a topic they enjoy. The later questions are equally weighted per mark and often easier to score on.
What to do instead
Write the mark allocation and a running clock target next to each question before starting. Move on when the target time is reached, even mid-answer, and return only if time remains.
Answering a 'discuss the risks' or 'critically evaluate' question with a list of risks but no assessment of likelihood, materiality, or mitigants.
Candidates equate listing with discussing. Higher-order command words require judgement about significance and about what the company should do.
What to do instead
For each risk point, add a short qualifier: how material it is in this scenario, whether it is already mitigated, and what action follows. Rank or group points by significance rather than listing them flat.
Failing to attempt or fully answer parts of a question because the topic feels unfamiliar, leaving whole sub-questions blank.
Candidates panic when a topic was not in their revision and believe a partial answer is worthless. Marks are awarded per valid point, and general insurance principles often generate several marks even without specific technical knowledge.
What to do instead
Never leave a sub-question blank. Apply first principles: what is the risk, who bears it, what data would you need, what could go wrong, what would you do. Write at least as many bullets as half the marks available, then move on.