Choosing the right structure

Before launching a survey, a structure needs to be created and then filled with respondents. The most important thing to understand is that the structure determines how results will be reported and what analysis options you'll have. There are two main approaches: a flat structure or a group-based structure. Which is best? It depends.

The key to choosing the right structure lies in focusing on the end result. How do you want to work with follow-up? Is it important to give different people access to different parts of the results? Are there clear groupings you want to divide the results by?

Flat Structure

In a flat structure, all respondents are placed in a single group. There are no summaries or subgroups such as country, region, location, customer segment, etc. To break down and analyse the results, those variables (e.g. country or location) must be included as background variables. The results are otherwise reported as one unified group.

🟦 Requires basic Excel knowledge and minimal manual handling when importing respondents.

Group Structure

Here, a hierarchy is created with summaries and subgroups that reflect the organisation’s needs – for example, splitting results by country, region, or site. This varies from company to company and is built based on what’s important to follow up on. A group structure enables deeper analysis, since the structural groupings can also be combined with background variables.

🟩 Useful to be comfortable in Excel and familiar with using filters to check for errors or duplicates before importing the file.

P.S. If something goes wrong during import, you can export a file showing what needs to be corrected in order to successfully complete the import.