Growing Your Business with Survey Data Analysis

Don’t let the term “survey data analysis” intimidate you – organizing and analyzing survey data so you can make actionable decisions to grow your business is easier than it sounds.

You already know that a well-executed survey can help identify areas for improvement in your business, but you may feel intimidated about the daunting task of analyzing your survey data. 

After all, pages and pages of data will not help you upgrade your business unless you know how to meaningfully analyze the data and draw conclusions. 

The good news is that you do not need a degree in data science to analyze your survey data like a pro!

This article explains how to execute survey data analysis, proving that the data you reap helps you draw conclusions about your target market and your industry. We broke down the process into four steps to make it easy to analyze survey data.

Step 1: Review your top questions and consider the responses

Start at the beginning by reviewing the goals you set for your market research survey. (If you are still in the planning phase, it will pay off later to carefully design your survey and set goals). Take the time to list out the top questions you want to answer during survey data analysis. This will keep you focused as you begin sifting through the data.

You also need to consider the types of responses your survey generated. Did you ask close-ended questions (yes/no or multiple-choice answers) or open-ended questions (fields with text entry allowing for a more elaborate response).

Close-ended questions let you generate empirical data that can be useful for drawing conclusions. Open-ended questions require careful review, but can reveal richer insights than empirical data can alone. 

Let’s consider a sample scenario where a business owner wants to know how they can improve their ecommerce business. A top research question in this example might be: “Are customers happy with the checkout experience?”

It is easy to find the answer to this question since it was asked directly during the survey:

 

Were you satisfied with the checkout experience?%Number
Yes74%148
No26%52

 

In this scenario, the majority of users are happy with the checkout experience. But what about the 5 respondents who are not happy? How can we use data to understand how to improve the checkout experience? For that, we need to dig deeper. 

Step 2: Review, filter, and cross-tabulate your data

If you are using a survey platform like Pollfish, you will have access to a powerful dashboard that allows you to view and filter your survey data. From the dashboard, you can filter and segment your survey results in real-time. For an advanced analysis, you should study it in a variety of formats, such as graphs, charts, and spreadsheets. You can use the latter to create crosstabs.

What are crosstabs?

Crosstab (short for cross-tabulation) is a special type of report that is used to explore the relationship between variables. It is essential in survey data analysis because it lets us segment survey data and examine responses for different segments. For example, we can examine satisfaction levels of the online shopping experience based on the subjects’ age, education level, payment type, etc. Pollfish provides crosstab functionality within the results dashboard, which streamlines the process for you.

Going back to our sample scenario, let’s see how we could determine which payment method is giving users the most problems. To do this, we need to crosstab the results to view payment types and satisfaction levels. 

We are looking at two sets of data:

  • The type of payment method used by respondents
  • Whether they were satisfied during the checkout process
Payment TypeSatisfied (yes)Unsatisfied (no)
Visa/Mastercard86%

(128)

6%

(3)

Apple Pay11%
(17)
6%

(3)

PayPal2%

(3)

88%

(46)

 

The crosstab report reveals that customers who used PayPal overwhelmingly expressed dissatisfaction with the checkout process, providing insight that something in the PayPal process is falling short.  

In the same way, you can apply crosstabs to examine the satisfaction levels voiced by other segments. Are people who shopped on their mobile devices happy with the checkout? What about an older segment of users compared to a younger one? 

Step 3: Understand the statistical significance 

Before drawing conclusions about your data and investing in changes to your business or website, you must crunch the data to understand if the results can be trusted. An important aspect of survey data analysis is assessing the statistical significance of your results. In the realm of data analysis, statistical significance is what helps us determine how accurate our data is. 

To do this, you need to consider these factors:

  • Sample size refers to the number of respondents in your survey. The larger your sample size, the more confident you can be about the results. 
  • Effect size describes the amount of difference between the data you are comparing. If you have a small effect size, you would need a larger sample size to understand if the difference is statistically significant (and worth acting on) or a result of chance.

In our online shopping example, the dissatisfaction voiced by PayPal users is significant and should be explored further. The percent of dissatisfied customers who used a credit card or PayPal is low enough that exploring this is unlikely to yield meaningful results, unless you can determine a third factor in this subset (for example, 100% of the dissatisfaction comes from mobile users). 

Step 4: Draw conclusions and create a plan for improvement

Now for the fun part! After you have organized, reviewed, and understood your data, it is time to draw conclusions and determine how this information can be used to improve your business. 

Go back to your original research questions. Sift through the data until you are able to answer each of these questions and draw conclusions. 

In some cases, a course of action will be very obvious. In our sample scenario, it is clear that this business owner needs to uncover issues in the PayPal checkout experience. 

It may be harder to understand why other segments feel unhappy with their shopping experience. 

For example, you may understand that those aged 65+ stated dissatisfaction but cannot find a clear relationship that explains why. In these cases, your open-ended questions may reveal insights that may help you interpret the dissatisfaction voiced by this segment. 

With your theories and conclusions in hand, create a plan for systematically improving each area of concern. In our example scenario, some changes to how the users move from the store to PayPal may improve the experience and overall satisfaction levels. Once you have made changes, you can understand their impact by running another survey and using your new data analysis skills to understand the change.

Pollfish makes it even easier

At Pollfish, we provide our clients with a dashboard that makes survey data analysis easier – you can review your responses in real-time and access visual data directly in the dashboard. You can view your data in classic mode or in a number of visual sources (thinks charts and graphs). You can also export your data for official reporting and set up cross-tabs. Ready to launch your survey?

Frequently asked questions

What is survey data analysis?

Survey data analysis is the process of organizing and reviewing survey data in order to draw conclusions and gain insights.

What are crosstabs?

Crosstabs, also known as cross-tabulations, are data tables that are organized in a way that allows a researcher to identify relationships between variables in survey data.

What is statistical significance?

Statistical significance is used in data analysis to understand how likely it is that survey results are accurate and not the result of random chance.

What is a sample size?

A sample size defines the total number of individuals who are chosen to participate in a survey or experiment.

What is the effect size?

The effect size is a numerical measure of the difference between two variables. The larger the effect size, the more confident a researcher can feel about the results of a survey or experiment.