Reporting that maximises insight
Michael Wagstaff • 22 September 2023
Best Practices in analysing and reporting survey data

Market research surveys are a valuable tool for businesses to understand their customers, test hypotheses and make informed decisions. However, the full value of this tool can only be realised when the survey data is analysed and reported correctly.
In this article we list six simple steps to ensure that survey data is both accurate and impactful.
1. Checking Data for Accuracy
- Data Cleaning: Before diving into analysis, ensure your dataset is clean. Remove or correct any outliers, duplicates or inconsistencies.
- Quality Control: Look for patterns indicating biased or skewed responses, such as participants selecting the same answer choice for all questions. Such responses may need to be excluded.
- Verify Sampling Methodology: Ensure the sample used is representative of the population you intend to infer about.
- Weighting the data : Apply weights if necessary so that the data are representative. Be wary of applying large weights to small samples as this will magnify any undue influence.
- If you're using an agency to do your survey make sure you discuss these steps in advance of the survey going out.
2. Computing New Variables
- Creating Indices: Combine multiple variables into an index to capture broader constructs (e.g., customer satisfaction based on various service metrics).
- Segmentation Variables: Compute variables that segment the audience into meaningful groups. For instance, use age and income brackets.
3. Making Inferences
- Statistical Significance: Always check if observed differences or trends in the data are statistically significant, especially when comparing groups.
- Causation vs. Correlation: Be cautious not to assume causation from correlations. Just because two variables move together doesn’t mean one caused the other.
- Multivariate Analyses: Sometimes, considering multiple variables simultaneously (e.g., through regression analysis) can provide more nuanced insights than univariate analyses.
4. Drawing Out Insight
- Ask the 'So What?' Question: After identifying a trend or pattern, delve deeper to understand its implication for the business.
- Synthesise with other data sources: Blend survey data with other datasets (like sales or website analytics) for richer insights.
- Benchmarking: Compare your results with industry benchmarks or past data from your own surveys to provide context.
5. Presenting Data for Maximum Impact
- Use Visual Aids: Graphs and charts often convey trends and patterns more effectively than tables or text.
- Executive Summaries: Start your report with a concise summary of the most crucial findings and their implications.
- Tell a Story: Structure your report as a narrative, guiding the reader through the most relevant points in a logical sequence.
- Avoid Jargon: Make sure your report is understandable to all stakeholders, not just those well-versed in research or your industry.
- Interactive Dashboards: Tools like Tableau or Power BI can help you create interactive dashboards. Stakeholders can dive into the data at their desired depth.
6. Feedback Loop
- Solicit Feedback: After presenting your findings, gather feedback from stakeholders to understand which insights were most valuable and why.
- Iterate: Use feedback to refine your next survey or the way you analyse and present data.
In conclusion, market research survey data has the potential to offer significant value to businesses, but only when approached with a rigorous and systematic analysis and reporting methodology.
By adhering to these best practices, you ensure that your insights are accurate, meaningful and actionable.