Analysis & Writing
Plan first. Fill in the gaps. Write one section at a time.
By this point you have your RQ, your five supporting documents, and a rough sketch of your thinking. The analysis section is where that becomes a 1,100-word argument.
Step 1: Plan your analysis section
Work through these three questions in order before you open a blank document.
What information do I need, and where will I get it?
List the specific pieces of evidence your argument requires. For each, identify which supporting document it comes from. If you cannot locate the information in your documents, you either need a different document or a different argument.
Which tools and theories will I use?
Choose the tools that genuinely fit your question (see Structure). Decide what each tool will do in your argument - what dimension of the RQ does it address? If two tools address the same dimension, drop one.
What structure will I use?
Choose your organising principle. The options are:
- For vs. against: advantages and disadvantages of the decision or strategy, especially useful for yes/no or should/shouldn't questions
- Internal vs. external stakeholder impacts: how different groups are affected differently
- Ethical / evaluative framing: "yes, the practice is justified because..." then "no, it is not because..."
- By business function: finance, operations, marketing, HR (if your RQ spans multiple functions)
- By analytical tool: one tool per sub-section, where each addresses a different aspect of the RQ (not the same point approached twice)
The test: could someone read just your headings and follow the logic of your argument toward an answer to the RQ? If yes, your structure is working. See Structure for more on this.
Write your actual sub-headings into a document. Then add rough notes under each heading - not prose, just bullet points: what evidence goes here, what the tool shows, what your conclusion for this section is. That is your plan. Now fill it in.
Description vs. analysis
This is the most important distinction in the whole IA. Description tells the reader what is there. Analysis tells the reader what it means - for the company, for the RQ.
"Nike's revenue grew by 10% in 2024. The company has a large marketing budget and invests heavily in sponsorships."
"Nike's 10% revenue growth in 2024 (Document 2) suggests that its pivot toward direct-to-consumer sales has not damaged overall volume, which weakens the argument that the strategy carries significant revenue risk."
Notice what the analysis version does: it names the evidence and its source, draws a conclusion from the data, and connects that conclusion directly to the RQ. Every paragraph in your analysis section should do all three.
You do not need to define or describe a tool before using it. The examiner knows what a SWOT analysis is. Name the tool, apply it to your specific company and data, and draw your conclusion. Explaining what Porter's Five Forces is wastes words and reads as description, not analysis.
Keeping the key concept visible
Criterion A rewards sustained integration of your key concept throughout the analysis - not just a mention in the introduction and conclusion. One practical way to do this is to bold the key concept word wherever it appears in your analysis. If you look at your draft and the word barely shows up, that is a signal to revise.
A useful writing habit: in the last sentence of each paragraph, loop back to the RQ and to the concept. Something like: "This suggests that [company]'s approach to sustainability is driven more by regulatory pressure than genuine strategic change, which directly bears on whether the decision was justified." That last sentence does double duty - it closes the paragraph and re-anchors the argument. Apply this as a guideline, not a formula: some paragraphs will close more naturally than others, and forcing it reads worse than skipping it.
Using your supporting documents
The examiner is looking for direct, specific use of the documents - not general knowledge. When you make a claim, cite the document and the specific data.
- Quote sparingly, paraphrase often. A short direct quote is strong evidence. A long block quote is a word-count problem. Use the number or the finding; summarise the reasoning in your own words.
- Reference by label. Refer to your documents as "Document 1", "Document 2" and so on - the same labels you use in the appendix.
- Use, do not describe. Do not write "Document 1 is an article about X." Just use it: "Revenue fell 12% in Q3 2024 (Document 1), which suggests..."
Getting started: write one BMT first
If you are staring at a blank page, start here. Pick the tool that feels most central to your RQ. Write a rough 200-300 word application of it using the data from your supporting documents. Do not worry about transitions or polish - just get the argument down.
Check the Sample IA to see how a tool application is structured: what the tool shows, what the specific data is, and how it connects to the RQ. Once you have one section working, the rest follow the same pattern.