- Daniel Joyner
- 3d
- 6 min read
Table of Contents:
Survey data can be incredibly powerful when it comes to brand measurement and category growth. Getting inside the minds of consumers to understand how they think about specific products, categories, industries, or brands can unlock countless opportunities. The most effective brands engage in continuous cycles of running and analyzing survey data, transforming that data into actionable insights, and then using follow-up surveys to fine-tune and optimize their performance.
But here’s the thing: To get the most out of survey data, companies need to understand how to get valuable data, clean and organize it, and, most importantly, turn that data into strategies that can be effectively activated.
Start by Designing Surveys That Deliver Real Insights
Of course, the first place to consider is the survey itself. When designing a survey, it is vital to consider how you will get meaningful response data. If questions are not well-organized or unintentionally bias the respondents toward certain answers, it can undermine the value of your survey data. While survey respondents aren’t likely to lie on a survey, it is common for respondents to simply not know the answer to a question or answer questions in ways that contradict their actual shopping behavior.
For example, if a brand wants to understand its lost buyers so they can attempt to recover them, they might discover that respondents are not aware that their spending habits have changed. And when probed, many respondents might go to the obvious answer that price is the largest factor in their decision-making, even if their behavior tells a different story.
Strong survey data comes from asking the right questions to the right audience in the right context. At Circana, we provide templated surveys based on tried-and-true methods that have worked for countless brands and retailers, as well as fully customized surveys that are designed to answer highly specific business questions. How and when our panelists are contacted is managed effectively to ensure that answers are unbiased and valuable. And most importantly, our panelists' answers are tied back to verified shopping data. By tracking how people shop alongside how they answer survey questions, we provide a more accurate view of sentiment.
Choosing The Right Goals for Your Survey
Another important element of strong survey analysis is ensuring that you are carefully considering what you want to study. One of the common pitfalls of survey data is when results are non-translatable to strategy. Your surveys should be built around answering the right questions for your brand so you can make decisive changes to your business strategies.
At Circana, some of the most common surveys we help with include:
New Trier Studies: These help brands understand how new products are performing and resonating with consumers.
Brand Recharge Studies: These help brands understand how and why they have lost shoppers and the methods that could help regain them.
Path to Purchase Studies: These dive deeper into the overall shopper journey that leads customers to their purchase decisions.
Brand Perception Studies: These help identify sentiment toward brands and their products
Leakage Studies: These look at how consumers' purchase behavior changes so that brands can understand why and how their consumers are shopping elsewhere.
By having a clear outline of what you are looking to discover, you can ensure that the data you will receive from your survey will be applicable to your business objectives.
Collecting, Organizing, and Analyzing Survey Data
In order to effectively analyze your survey data, you need to harmonize it with other datasets. Great survey analysis requires a combination of behavioral and attitudinal data. The first and most important step is organization. You will need to be able to compare what you know about the respondents against how they responded. Their purchase data, demographic/audience data, and other unique attributes should be organized so you can more effectively see how different audiences responded to questions.
Cross-tabulation is a great way to start. It can help break down your data based on demographics. If you are already working with target audiences, this can help show how those groups are responding to your brand, marketing efforts, or your products. Looking for differences between groups can also help build your data story.
When you can see the major themes of your respondents, you need to start addressing the “whys”. Qualitative data, including open-ended questions, can help provide context for why different groups of respondents answered other questions the way they did. Of course, one of the most valuable insights you can have is verified purchase data. If you know how respondents are shopping, this can help support more accurate insights and conclusions. While survey data might show that consumers are interested or disinterested in a product/service, their purchase data can tell a different story. Having access to shopping data is always beneficial as it can help ensure that your insights are based on what is truly driving shoppers' purchase decisions.
Apply Survey Insights, Iterate, and Dive Deeper
Surveys play a pivotal role in long-term brand strategies. It’s a never-ending cycle of gaining insights, applying them, and using performance data to inspire future surveys. While surveys are effective at making singular decisions about launching new product varietals or learning about brand resonance, the best approach to survey data is to use survey data to both support immediate next steps and long-term brand equity analysis. While templated surveys can help answer common questions, working with survey writing & analyzing experts can unlock customized surveys that allow brands to dive deeper and discover more.
Frequently Asked Questions About Survey Analysis
What Is the Difference Between Quantitative and Qualitative Survey Data?
Surveys garner both quantitative and qualitative data. Quantitative data refers to anything with clear numerical value. This data can easily be analyzed via cross-tabulation. Groups can be created within the survey respondents, and their answers can be compared to other groups to identify key trends. Qualitative survey data is non-numerical and can include opinions, observations, and open-ended question responses. This data is more difficult to organize, but it provides deeper insight into the opinions of respondents. Generally, an effective method for analyzing these two data types is to organize quantitative data to discover trends, and then use qualitative data to gain a deeper understanding of those trends.
How Do You Analyze Different Types of Survey Data?
Surveys create a variety of different types of data.
Closed-ended questions refer to situations where respondents have to choose from provided answers. This can be a simple yes or no, or it can be a multiple-choice question.
Open-ended questions allow respondents to provide their own answers to a question. While open-ended questions are more difficult to analyze holistically, they can provide deeper insights or a multi-dimensional view of why respondents answered close-ended questions the way they did.
Nominal data includes categorical data that respondents provide. For example, the marital status of respondents can be used to create groupings.
Likewise, ordinal data is also categorical but can easily be arranged in a sequential order, like agreement scales where respondents can choose an option on a clear spectrum, including “very satisfied” to “very dissatisfied”.
Interval data uses numerical values where, for example, respondents can rate their opinion on a scale from 0-10.
To analyze this data, you need to organize signals from nominal, ordinal, and interval data in order to identify key trends between different groups. Open-ended questions can then be used to understand the potential causation behind correlations. For example, if a majority of unmarried respondents do not buy a product, open-ended questions can provide potential reasoning. Perhaps the product’s size might be a better fit for a family, and therefore, unmarried respondents find the product too expensive, or have frequently found themselves throwing it out if it has a shelf life. From these insights, a brand could start testing the concept of a smaller variety of the product in order to capture the underserved group.
What Are Common Survey Data Analysis Methods?
There are many different methodologies that can help analyze responses from surveys.
Cross-Tabulation: This method compares two or more variables in order to determine a corresponding relationship. For example, respondents can be grouped by age demographics, and then their answers to questions can be compared to see if there are clear trends within these audience segments.
Cluster Analysis: This method groups different data points together in order to see similarities. It is an effective way to segment respondents.
T-Test: This statistical test determines if there are significant differences between the means of two different groups. For example, respondents could be surveyed twice before and after a marketing campaign. A T-test could help support whether or not a marketing effort had a clear effect on brand awareness or positioning.
Analysis of Variance (ANOVA): While a T-test looks at differences between two groups, analysis of variance compares the means of three or more groups to check for statistically significant differences. For example, ANOVA could show that one or two generational groups are significantly different from the rest in regard to satisfaction.
