Writing the executive summary of a market research report

What is the executive summary for?

The executive summary is the part of a report that travels. It is forwarded to people who never saw the brief, pasted into board papers and read on a phone between meetings. For many readers it is the whole report, so it must stand on its own: no references to "the chart on page 14", no undefined abbreviations, no finding that only makes sense after reading the method.

It is not an introduction, which explains why the study was done, and not a table of contents written as prose ("Chapter 3 covers the market…"). It also differs from a topline report, which lists first results soon after fieldwork with little interpretation. The summary interprets: it tells the reader what the evidence means for their decision.

Which structure works best? Answer first

Decision-makers read from the top and stop once they have enough. The structure below puts the most valuable sentence first and the supporting detail after it.

Consultants know this logic as the pyramid principle: the conclusion on top, the supporting arguments beneath it, the evidence under each argument. If the client asked an open question rather than a yes-or-no one, block 1 becomes the single most important insight the study produced.

Five rules for a summary people act on

1. Lead with the answer, not the method

Opening with "This report presents the results of a study conducted in…" spends the reader's best attention on the least useful information. State the answer, then compress the method into one sentence: who was asked, how many, how and when.

2. One finding, one sentence, one number

Each finding should be a complete sentence that would work as a headline, carrying the figure that proves it, with period and source. "Adoption is growing" is an opinion; "33.2% of Dutch enterprises used AI in 2025, up from 23.1% a year earlier (Eurostat)" is a finding.

3. Say what each finding means

After the fact comes the "so what". A number without an implication leaves readers to draw their own conclusions, and different readers will draw different ones. Keep implication and fact visibly apart, so that someone can dispute the first without doubting the second.

4. Be honest about confidence

Give the sample size and, for survey figures, the margin of error. Say when findings are qualitative and show reasons rather than shares. Name what the study could not measure. A summary that admits a limit persuades more, not less, because readers then trust the rest.

5. Nothing new, nothing missing, nothing different

Every claim in the summary must appear in the body, with the same number rounded the same way. No new data enters at this stage, and no finding that changes the answer is left out. A mismatch between page one and page twenty is the quickest way to lose a board's confidence.

Before and after: a hypothetical example

Example: a hypothetical Irish software company sells AI-based document tools to mid-sized firms. Its management asks whether to prioritise the Netherlands or Bulgaria for expansion in 2027. The study combined Eurostat data with 20 interviews with IT managers in both countries. Below is a first draft of the summary, then the rewrite.

Before: "This report presents the findings of a study carried out in September and October. The research combined desk research and interviews with IT decision-makers. Artificial intelligence is a growing area across Europe, and the results show interesting differences between the two countries. Both markets have potential, although some challenges were identified. Further research is recommended."

After: "Prioritise the Netherlands in 2027 and treat Bulgaria as an option for 2028. Adoption decides the order: according to Eurostat, 33.2% of Dutch enterprises with 10 or more employees used at least one AI technology in 2025, against 19.9% across the EU and 8.6% in Bulgaria. The Dutch share rose by 10.1 percentage points in a single year, so buyers there already compare suppliers rather than ask whether AI is worth trying. Bulgaria's share is small but has more than doubled since 2023, when it stood at 3.6%. In the interviews, Dutch IT managers raised data location and GDPR before price, while Bulgarian managers asked mainly about support in their own language. The interviews show reasons, not shares, so a buyer survey is needed to size demand in each country. Next step: a Dutch pilot with three reference customers, and a review of the Bulgarian option when Eurostat publishes its next figures."

What changed between the two versions:

The rewrite is longer, but every sentence does work. For business-to-business questions like this, B2B research with decision-makers supplies the interview layer, and a market entry study adds regulation and sales channels.

How long should the executive summary be?

One page for a report of up to about 30 pages, and two pages at most for longer ones; in words, roughly 250 to 600. In a slide report, use one slide for the answer and key findings and a second for risks and next steps. If the summary will not fit, the usual cause is too many findings: rank them by their effect on the decision and keep three to five.

Charts are optional. A single chart that carries the central finding can help, provided it has a headline and a source line; a page of six small charts turns the summary back into a report.

When should you write it, and who should check it?

Write the summary when the findings chapters are stable. You can draft the one-sentence answer early as a working hypothesis, but expect to rewrite it. Our guide on how to write a market research report shows where the summary fits in the overall process, and the annotated market research report example shows a finished summary in context.

Then run three checks:

• Number check: every figure matches the body, with the same rounding and source.
• Cold read: a colleague outside the project reads only the summary and explains the recommendation back to you.
• Plain-language check: short sentences, no jargon, abbreviations spelt out. This matters when many readers work in a second language, as in most international teams.

Mistakes that make readers skip the summary

• Methodology first. Two paragraphs on fieldwork before any finding.
• Adjectives instead of numbers. "Significant growth", "strong interest", "considerable potential".
• Every finding included. Twelve bullets of equal weight leave the reader to rank them.
• Hedging everything. "May", "could" and "potentially" in every sentence hide which findings are solid.
• Recommendations without owners. "Consider improving the offer" gives nobody anything to do.
• AI-drafted summaries left unchecked. Language models summarise fluently but can merge findings or round numbers differently from the tables; check every figure. Our article on market research with AI covers this risk in more depth.