Surveys are a key tool for gathering people’s opinions. To better segment these opinions, we often collect objective socio-economic data such as income, age, or gender alongside subjective responses. While opinions are inherently personal, when collected from representative samples, they can be averaged to accurately describe the views of broader population groups. These results can then be used to benchmark companies, products, and services.
French survey to benchmark the performance of French sports stores
Although average scores are commonly used to represent group opinions and compare research targets—such as products, services, or companies—the „experience horizon“ of respondents must be taken into account. Different user groups, such as customers of different companies or users of different products, often have varying experiences.
For example, a customer of an insurance company offering a 24-hour hotline will naturally have a different frame of reference than a customer whose insurer only provides hotline support for 8 hours per day. Likewise, the way respondents interpret evaluation scales (e.g., 1 to 10) can vary significantly. Therefore, customer survey benchmarks provide important indications of potential performance but cannot claim absolute objectivity.
Quality of Survey Design is Critical
Survey quality hinges on two main factors: questionnaire design and respondent sampling.
The clearer and more precise the questions, the better the quality of the answers. However, there is no foolproof method to ensure every respondent interprets a question the same way. Good survey design aims to minimize, though not eliminate, sources of misunderstanding.
High-quality surveys also depend on careful selection of respondents. Participants must meet specific screening criteria before they can take part. Those who do not qualify are excluded to maintain the survey’s relevance and reliability.
For example, a typical screening criterion in a CGG study might be whether respondents are customers of a particular company or have used a certain product or product category recently.
A French sports store survey included a screener question:
„Have you visited a sports store in the past 12 months?“
Sample size is another crucial factor in survey quality. There is no single „correct“ sample size that fits all needs, and a perfectly representative sample is only theoretically possible with the entire population. However, by applying logical assumptions about the distribution of results (usually assumed to be normal distribution), along with calculations for standard deviation, confidence intervals, and margin of error, an appropriate sample size can be determined.
CGG Surveys Typically Feature Samples of Over 1,000 Respondents
At Consumer Guidance Group (CGG), we aim for sample sizes of at least 1,000 respondents in population-wide surveys. For surveys involving smaller populations (niche groups), proportionally smaller sample sizes are acceptable.
Our surveys accurately reflect the opinions of users or customers of comparable products, services, or companies. By employing consistent scales and rigorous methodologies, they allow meaningful benchmarking across different research targets—helping users and consumers make more informed decisions.