The Institute and Faculty of Actuaries sets an exam about building a model, and gives most of the marks for writing about it. From A Guide to Modelling Practice (IFoA, February 2025): "The communication aspects are the main part of the exam and these account for around 80% of the marks. There are relatively few marks for the techniques used in the models." The same guide notes that in Paper 1 there are often 30 marks available out of 100 for doing the spreadsheet modelling and the checks at all.
That inverts what most applicants believe about this profession. is two papers of 3 hours 20 minutes each, and the model itself is the minority of the marks. Next to it sits Communications Practice, a whole subject whose stated aim is communicating actuarial concepts "to recipients without technical expertise". Both sit inside the qualification, the first one, not Fellowship.
So the profession has already decided: explaining the work is the work. Almost nobody teaches it before you get the job. That is what this course is for.
The IFoA names its readers
names them, and it is unusually specific about what each one needs. In Paper 1 you provide "a clear for the following individuals: a fellow student you should assume has about the same level of knowledge as you; and a senior actuary who will know the actuarial aspects and will need an overview of the actuarial, technical and assumption elements. They will not need the Excel details." One document, two readers who want different things from it. In Paper 2 you "provide a summary of the whole project which will be for the senior actuary only": one document, one reader.
| Reader | Assume they know | They need from you | They do not need |
|---|---|---|---|
| A fellow student | roughly what you know | every step, in order, well enough to rerun and correct your work without opening it | the subject explained from first principles |
| A senior actuary | the actuarial content, cold | the approach, the assumptions, the results and what they mean | "the Excel details" |
The pass criterion carries both readers in one sentence: Paper 1 wants "a clear audit trail, including checks, which could be followed by a senior actuary and would enable the model to be worked on and corrected by a fellow student". It is not even the same model twice: Paper 1 documents the one you built, Paper 2 has you review and extend someone else's before summarising it. A later lesson in this course, on the audit trail and the summary, is how you actually write each document. The point for now: the reader is a variable, and much bad actuarial writing is good writing aimed at the wrong person.
Your year-one cast
The job gives you more readers than the exam does. In a first year they come down to six recognisable people, each holding a different decision.
| Who | The decision in their head | What the first sentence must give them |
|---|---|---|
| Your reviewer or checker | can I sign this off? | what changed, what you checked, what you did not |
| Your line manager | is this ready to go out, and does it need me? | the answer, then the one thing that could bite |
| A senior actuary or partner | do I believe this number enough to put my name on it? | the result, the basis it sits on, the key judgement |
| A trustee or client | do I need to do something? | what happened, in their words, and whether it matters |
| An underwriter or claims lead | does this change what I do on Monday? | the number that moves their decision, and by how much |
| A regulator or auditor | can this be evidenced? | what was done, when, by whom, and where the trail is |
Very little of that list is about being technically impressive, and hardly any of the six want your method.
What actually changes between readers
Four things, and you set them before you start writing.
Jargon budget. With a fellow analyst it is unlimited and saves everyone time. With a trustee it is roughly zero, and every technical term you spend has to buy something.
Where the caveat sits. For a reviewer, caveats sit next to the number they qualify, because their job is to test them. For a client, a caveat buried in paragraph nine has not been communicated at all. Put it where the decision is made.
Length. Length is a claim about how much of the reader's time your work deserves. The guide is blunt about what happens when you get that wrong: "Eight pages of densely typed text in 8 point font will not produce an easy to follow or intelligible audit trail."
The ask. Every piece of work ends with something the reader is meant to do: approve, decide, note, or nothing at all. If you cannot name it, you are writing to feel busy.
Three questions, before a word
- Who reads this, by name? Not "the client": the pensions manager who will forward it to nine trustees.
- What decision are they making, and when? A note that lands after the decision is a filing exercise.
- If they read only the first sentence, do they still decide correctly? If not, rewrite the first sentence. Many readers genuinely read only that one.
It is marked long before CP2
APR, a UK actuarial consultancy that publishes its own graduate process, runs an written test covering, in its wording, mathematical reasoning, probability and statistics, programming (although no specific language knowledge is required), insurance industry knowledge, and written communication. The generic marking criteria for graduate written exercises, as set out by targetjobs (21 June 2023), are clarity of ideas and structure, extracting the key points from data, explaining processes in simple terms, appropriate tone for the audience, and spelling and grammar. Nothing on that list rewards sophistication.
Check your understanding
Your Paper 2 summary for the senior actuary opens: 'Cell D9 applies the 1.08 development factor to column C, copied down to D12.' What has gone wrong?
The same caveat has to reach two readers: the reviewer checking your work and the client acting on it. Where does it sit in each document?
You finish a draft and realise you cannot say what its reader is meant to do with it: not approve, not decide, not even note. What has gone wrong?
Your note's first sentence reads: 'This paper sets out the methodology adopted for the quarterly review.' You know its reader often stops after one sentence. What should you do?