How to write an analyst-grade market research report

An analyst-grade research report is one whose reasoning a sceptical reader can inspect: every claim is traceable to a source or to a stated inference, and the methodology is explicit enough to be argued with. That is the whole standard. It is not about page count, chart density, or the confidence of the prose — those are the things that imitate rigour when rigour is absent.

Here is the structure that credible market reports share, and the sourcing discipline that makes them survive a hostile read.

Start with one question

Before anything else, write the question the report answers in a single sentence. Not a topic — a question.

• Topic: "The European fleet telematics market."
• Question: "Is the European fleet telematics market large enough, and fragmented enough, to support a new entrant selling to operators with fewer than 50 vehicles?"

The topic version produces a report that describes things. The question version produces a report that decides something. If you cannot compress your question into one sentence, you have more than one report, and writing them as one will serve neither.

The nine-part structure

Sections 1 and 8 are where reports usually fail. The executive summary gets written as a table of contents in prose ("this report examines…") instead of as the answer. The implications section gets dropped for time, which leaves the reader holding a description of a market and no idea what to do about it.

How should you handle sources and numbers?

Give every figure in the report a provenance class. Three classes are enough, and they should be visible in the text or in a marker beside the number:

• Reported — the number appears in a named source. Cite it, with the date. A 2023 figure quoted in 2026 is a 2023 figure, not a current one.
• Derived — you calculated it from reported figures. Show the calculation, at least in the appendix. "Derived" without the arithmetic is just "estimated" wearing a better coat.
• Estimated — you modelled it from assumptions. Name the assumptions and give a range rather than a point.

Mixing the three silently is the single fastest way to lose a technical reader. Once someone discovers that a confidently stated number was actually a guess, they will — correctly — start treating every other number as one.

The date problem

Market figures decay. A report that cites 2022 data without saying so implies currency it does not have. Two habits fix this: put a cut-off date in the method section ("sources current as of 12 August 2026"), and attach the vintage to individual figures whenever it is older than the cut-off. It costs four words and buys the reader's trust for the rest of the document.

Say when you do not know

Some markets have no reliable public data. Private company revenue, early-stage category sizes, regional splits in fragmented industries — these frequently cannot be sourced. A rigorous report says so and explains what it did instead. A report that never says we don't know is not more rigorous; it is merely more confident, and confidence without evidence is exactly the failure mode readers are scanning for.

The credibility of a research report is set by how it handles its weakest claim, not its strongest. One unsupported assertion presented with the same certainty as a well-sourced one drags the whole document down to its level.

Writing the executive summary

Write it last, and write it as though it will be the only page read — because it usually is. A working summary has four moves in about 250 words:

1. The answer. One or two sentences that directly answer the question from section 2. Not "the market is complex" — the actual answer.
2. The three findings that support it. Each a sentence with a number in it.
3. The main uncertainty. What would have to be true for the answer to be wrong, and how likely that is.
4. The recommended action. Specific enough that a reader could act on it, or explicitly decline to.

If your summary could be pasted into a report about a different market with only proper nouns changed, it is not a summary. It is throat-clearing.

What makes charts help rather than decorate?

Every chart in a decision-grade report should be defensible in the same way a number is. Three rules cover most of it:

• The title states the finding, not the variable. "Mid-market share has fallen 11 points since 2023" beats "Market share by segment, 2020–2026." The reader gets the point before parsing the axes.
• The source and provenance class sit under the chart. A modelled chart and a reported chart look identical; only the caption can distinguish them.
• If the chart does not change what a reader believes, cut it. Volume of exhibits is not evidence of work, and every chart the reader skips lowers their attention for the next one.

A self-check before you send it

1. Can I state the question the report answers in one sentence, and does the executive summary answer that question?
2. Does every number carry a class — reported, derived, or estimated?
3. Is there a section a hostile reader would attack first, and have I strengthened or acknowledged it?
4. Have I said, somewhere, what I could not establish?
5. Would the implications section be useful to this specific reader, or is it generic advice that could go in any report?

Reports that pass those five checks tend to survive their meeting. Reports that fail them tend to get one polite question that nobody can answer, after which the recommendation quietly dies.

If you want the structure but not the fortnight, this is exactly the shape of document the market sizing and research work on the bench produces — sourced, classed, and delivered as both a web page and a PDF.