Most insight reports fail at the last step.

The research is solid. The methodology is sound. The analysis is thorough. And then it lands in a deck that walks executives through the methodology, explains how respondents were recruited, and then presents findings as a list of bullet points that nobody reads after the meeting.

The problem is structural. Research is typically presented as a documentation of what happened in the study. What stakeholders need is a brief on what it means for their decisions. Those are different documents with different structures, different languages, and different roles for the data.

This guide is about writing and presenting the second kind of document.

Start with the Business Question, Not the Methodology

Every research project starts with a business question. “Should we expand into Tier 2 cities?” “Why are trial users not converting?” “Which product concept resonates most with our core customer?”

Your research presentation should open by restating that question, then answering it as directly as possible. Not “here is what we found.” But: “Based on this research, the answer to the question we started with is X, and here is why.”

This sounds obvious. In practice, almost no research presentations are structured this way. They open with methodology, then present findings theme by theme, then close with implications. The implications, the part that matters are buried at the end when stakeholders have already formed their own interpretations of the findings.

Lead with the answer. Then show the evidence.

The Pyramid Structure for Research Presentations

The pyramid principle, developed by Barbara Minto at McKinsey, organises communication from the top down: answer first, then supporting arguments, then evidence. It is the natural structure of a memo written for a busy executive. It is also the natural structure of a good research presentation.

Applied to research, the pyramid looks like this:

1. The headline finding: one sentence that answers the core business question
2. The three to five key insights that support or qualify the headline
3. The evidence for each insight: data, quotes, survey results

Most research reports do the reverse: evidence first, findings second, headline buried at the end. Stakeholders who are reading quickly miss the headline. Stakeholders who only scan the deck miss everything except the executive summary.

What Goes in the Executive Summary

An executive summary for a research report should be exactly one page or one slide. It contains:

• The research question and the answer to it
• Three to five key findings in one sentence each
• The recommended actions that follow from the findings

Nothing else. Not sample size. Not methodology. Not fieldwork dates. Those belong in an appendix for anyone who needs to evaluate the research quality, not in the executive summary that your CMO is reading on their phone between meetings.

Writing Insights, Not Observations

There is a difference between an observation and an insight. An observation describes what happened in the data. An insight explains what it means.

Observation: “62% of respondents said they were satisfied with the checkout process.”

Insight: “Satisfaction with checkout is high among users who discover the product through social media (71%) but significantly lower among users who came through paid search (49%), suggesting that paid search is attracting a different audience with different expectations.”

Observations tell you what. Insights tell you why it matters and what to do about it. Research reports filled with observations require stakeholders to do the interpretive work themselves, which they rarely do or do incorrectly without the full context of your data.

Write insights, not observations. Every data point in your report should be there because it supports an insight, not because it came out of your analysis.

Using Verbatim Quotes Effectively

Qualitative research produces verbatim quotes that can do something charts and percentages cannot: make findings feel real and human.

Used well, quotes anchor abstract insights in concrete human experience. Used poorly, they are selected to confirm the researcher’s pre-existing interpretation and read as cherry-picked anecdotes.

Best practice for using quotes in presentations:

• Use quotes to illustrate insights, not to introduce them. The insight should be stated as a finding first, with the quote as supporting evidence.
• Use multiple quotes when possible, especially if they come from respondents with different profiles. One quote is an anecdote; three quotes with different backgrounds are patterns.
• Include quotes that complicate or challenge your main narrative. Research that presents only confirming evidence looks suspect to experienced stakeholders.
• Attribute quotes with relevant demographic context: “Female, 34, Tier 2 city” is more useful than “Respondent 7.”

Handling the “So What?” Question

Every finding in a research presentation should be able to survive the “so what?” question from a skeptical stakeholder. If you cannot connect a finding to a specific business implication or decision, it should not be in the main body of the report.

The practical test: for each finding, ask “Given this finding, what should the business do differently?” If you cannot answer that question, either the finding needs deeper analysis or it belongs in the appendix.

This is not about forcing research to tell people what they want to hear. It is about doing the analytical work to connect data to decisions, which is the actual job of an insight professional. The data is not deliverable. The decision-relevant interpretation of the data is the deliverable.

Visualising Quantitative Data in Research Reports

A few rules for charts in research presentations:

• Use bar charts for comparing groups or categories. Use line charts for trends over time. Use scatter plots for relationships between variables. Do not use pie charts for more than two or three segments, they are harder to read than bars.
• Label data points directly rather than relying on a legend whenever possible. Eyes should not have to travel from chart to legend and back.
• Highlight the finding in the chart title, not just the variable name. “Consideration Drops Sharply in 35-44 Age Group” is a better chart title than “Brand Consideration by Age.”
• Do not show every data point. Show the data that supports the finding. Other data lives in the appendix.

Presenting Qualitative Findings Without Sample Size Anxiety

Qualitative research findings are sometimes undermined in presentations by stakeholders asking “but how many people said this?” treating qualitative data as if it were survey data with an insufficient sample.

The response is to reframe what qualitative data does. Qualitative research does not measure the prevalence of a view. It explores and explains why people hold views, what drives behaviour, and what the nuances of an experience are. These are questions that a survey with 500 respondents cannot answer.

Explain thematic saturation if challenged: qualitative sample sizes are determined by when no new themes emerge, not by statistical power calculations. This is not a limitation of qualitative research, it is a design choice that makes it appropriate for different questions than quantitative research answers.

Fielding Questions in the Debrief

Live research debriefs generate questions that you may not have anticipated. The best preparation is to know your data well enough to distinguish between three categories:

• Questions you can answer directly from the data: answer them
• Questions the data cannot answer: be direct that this is out of scope and, if relevant, recommend a follow-on study
• Questions where the data is ambiguous: present the range of possible interpretations and recommend how to resolve the ambiguity

Avoid the temptation to speculate beyond the data. Researchers who speculate in debriefs become associated with unreliable findings. Researchers who are honest about the limits of their data become trusted advisors.

For guidance on how to analyse qualitative data rigorously before presenting it, see How to Analyse Qualitative Data: A Step-by-Step Approach.

Related Reading

• How to Analyse Qualitative Data: A Step-by-Step Approach
• How to Measure Brand Awareness: Methods, Metrics & Benchmarks
• How to Conduct a Brand Tracking Study
• In-Depth Interview (IDI) Guide: Templates, Tips & Best Practices
• What is a Research Brief?
• Sample Size for Qualitative Research: How Many Respondents Do You Need?