A good survey report is one that turns the raw data from a questionnaire into insight a reader can actually use — and act on. The results from your survey have landed. Now they have to be transformed into something more than raw data. The results need to be brought to life. But first, the data has to go through the analysis machine and be turned into a report that is easy to understand and at the same time something the organisation can act on.
This article gives you the foundational principles of report design — built on three questions you should answer before you open the report template, the choice between a classic report and a dashboard, the right level of detail for your audience, and the visual principles that decide whether your readers actually engage with the result.
Highlights
What is a survey report
A survey report is the analytical output that turns raw questionnaire data into a document or dashboard the organisation can read and act on. Effective survey reports are designed for the specific reader, the specific use case and the specific level of subject-matter knowledge the audience already has. Three questions — who reads it, what they will use it for, and what they already know — decide most of the report's structure and detail level before any chart is drawn.
The 60-slide report nobody opened twice
The survey closed three weeks ago. The data is clean. You've built a 60-slide report with every cross-tab, every breakdown, every chart the platform offers. You present it to leadership on Monday. Wednesday, nobody can remember the headline finding. By next quarter, the report sits in a shared folder nobody opens. The work wasn't wasted because the data was bad — it was wasted because the report was built for the data, not for the reader. A good report inverts that: built for the reader first, the data second.
Understand the tasks and the audience
Before you start building a report, it is a good idea to think carefully about who will read and use the report — and what they will use it for. A survey often starts with a need to investigate certain areas. How a report should be designed, and what it should contain, depends on the answers to three questions.
Who will read and use the report?
Is the recipient an individual or a larger group? Is it leadership or employees? If you have a small number of users you know well, it is often easier to make a sharp, focused report than if there are many users — in which case you have to make sure everyone can understand it.
What will they use the report for?
Of course the users will read the report to get more insight. But it is worth going one step further. Do they need strategic insight? A tool for making strategic decisions? Or should the report function as a reference document they can come back to? Each of those use cases produces a different report.
What is their level of expertise?
Are the users used to working with survey results, or is it new to them? Do they know much about statistics, or does most of it need to be explained from scratch? And how well do they already know the subject the survey is about? If the users already know a lot about the topic, the report doesn't need to contain as much background as it would if the topic is unknown to them.
Once you have figured out what the report will be used for, who will use it, and what they already know about the topic, you are ready to put on the reader's glasses and design the report around what works best for them.
Report vs. dashboard — which format fits?
A question that often comes up early in the reporting phase is which format to present the results in. In Surveyxact, there are two main formats: the classic report and the dashboard. Technically the two formats work fairly similarly. They can contain many of the same elements, and they will always pull updated data from the measurement. In many cases, the choice is partly a matter of taste, but there are real differences in functionality and user interface — and a few principles worth starting from.
When is a classic report best?
When is a dashboard best?
Include only the necessary information
Is your report very detailed? Then it can quickly become heavy reading. If you present too much data in the report, the reader will have a hard time finding the important and interesting results. Results and findings from surveys can often be presented in many different ways with different levels of detail.
There are countless ways to present a result, but an important question is: how detailed should it be? A result should not be presented in overwhelming detail, but it should still contain enough detail for the reader to understand the meaning of, for example, a figure or analysis. So finding the right level of detail for what we are presenting is a bit of a balancing act.
In some contexts, "the more the better" is true. But when it comes to reports based on a survey, that is not the case. The simpler you can convey the results, the easier it is to get the message across. But as always — it depends on the reader and the purpose.
Frequency vs. average
How should you present results from questions based on a rating scale? For example a 5-point scale from "strongly agree" to "strongly disagree", "very satisfied" to "very dissatisfied", or "very important" to "not important"?
If you show the percentage shares for all five response categories, that is called showing a frequency distribution. If instead you condense it down to a single average, you reduce the number of figures from 5 to 1. The same applies if you show it as an index from 0–100, or simply the share that answers, say, 4 and 5 (typically the two most positive response options).
What makes the most sense to show depends on the reader and the purpose.
Rule of thumb
If you are presenting differences between several groups, you generally want to reduce the level of detail. The point of breaking the data down is to surface differences and patterns across groups of respondents. The lower the detail level, the easier it is to spot those differences and patterns.
Remember the data minimisation principle
GDPR rules state that the data we process about the registered persons (the respondents) must be strictly necessary to fulfil the purpose of the survey. Avoid "nice to know" and stick to "need to know".
In other words: we should only collect and present data we actually need. So limit your reporting to what is necessary — both because it produces better reports and because it is what the regulation requires.
A clear and logical structure
A report that looks professional and structured comes across as credible. Which structure and sequence is most appropriate depends on what the report will be used for:
- Strategic insight
- Tools for making strategic decisions
- Reference document
- Whether the questionnaire follows a model or not
- What should be in the report itself, and what can sit as an appendix
In SurveyXact, there are useful tools to help you build a good structure — including table of contents, headings, page numbers and page breaks.
Layout and design
Adopt a visual mindset. Our brains process images and figures before — and faster than — text. Be aware of:
- We read from left to right and from top to bottom.
- Be consistent throughout the entire report.
- Font and font size.
- Placement of elements.
- Use of colour.
- Assess every figure, page and report/dashboard against readability.
Which colours should you use?
Which colours you choose for your reports is partly a matter of taste. Colours can evoke specific feelings or symbolise things like hope, growth or warning. So be conscious of the colours you use.
That said, a few general principles consistently work:
- For the same type of data, use one colour across all bars and figures of that type.
- If you use colours to convey different types of data, be consistent throughout.
- If 2023 gets the colour blue and 2022 gets the colour yellow, that pairing should run through the entire report.
- Make sure there is enough contrast between categories.
"The best reports aren't the ones with the most data — they're the ones the reader can act on within five minutes of opening them. Every chart that doesn't help the reader make a decision is a chart that gets in the way of the ones that do."
— Morten Holm Therkildsen, Account Manager and Survey Expert, Surveyxact, Ramboll
Numbers backing this article
Frequently asked questions about survey reports
What makes a good survey report?
A good survey report is one that turns raw questionnaire data into insight the reader can actually use and act on. The three things that decide quality are: knowing who the reader is, knowing what they will use the report for, and knowing what they already know about the topic. Get those three answers settled before you build the report, and the rest of the design follows naturally.
When should you use a dashboard instead of a classic report?
Use a dashboard when the reader needs an overview, will make their own filtering and navigation choices, will primarily use it online, and the underlying measurement is ongoing with results updating regularly. Use a classic report when you want to tell a story with a beginning and an end, the reader doesn't need to make choices, the report needs to be printable or exportable to PDF/Word/PPT, and the underlying measurement has already closed.
How detailed should a survey report be?
As detailed as the reader needs, and no more. If you present too much data, the reader will have a hard time finding the important findings. The simpler you can convey the results, the easier it is to get the message across. The exception is when you are presenting differences between several groups — there, you generally want to reduce detail (use averages or indices) so the patterns across groups are easier to spot.
Should you show frequency distributions or averages for rating-scale questions?
It depends on the reader and the purpose. Frequency distribution (showing the percentage for each of the five response categories) gives the full picture but requires more cognitive effort. Average, index (0–100) or top-2-box share reduces the result to one number, which is easier to read but loses nuance. As a rule of thumb: use frequency distribution when nuance matters; use a single summary number when you are comparing groups or tracking change over time.
How does GDPR affect what you report in a survey?
GDPR's data minimisation principle states that the data processed about respondents must be strictly necessary to fulfil the survey's purpose. In practical reporting terms: avoid "nice to know" data, stick to "need to know". Limit both the data you collect and the data you present to what is required — both because it produces better reports and because it is what the regulation requires.
Key takeaways
