So, you're looking to get some feedback from your customers. That's great! However, surveys and data collection are more complex than simply asking customers how they feel about your company.
Obtaining customer feedback is difficult. You need strong survey questions that effectively derive customer insights. Not to mention a distribution system that shares the survey with the right customers at the right time.
And, even if you're successful that effort will be all for not if you're not sure what you've gathered. After all, survey data doesn't just sort and analyze itself. You need a team dedicated to sifting through survey results and highlighting key trends and behaviors for your marketing, sales, and customer service teams.
In this post, we'll discuss not only how to analyze survey results, but also how to present your findings to the rest of your organization.
How to Analyze Survey Results
1. Understand the four measurement levels.
Before analyzing data, you should understand the four levels of measurement. These levels determine how survey questions should be measured and what statistical analysis should be performed. The four measurement levels are nominal scales, ordinal scales, interval scales, and ratio scales.
Nominal scales classify data without any quantitative value, similar to labels. An example of a nominal scale is, "Select your car's brand from the list below." The choices have no relationship to each other.
Due to the lack of numerical significance, you can only analyze mode from this type of scale. You can keep track of how many respondents chose each option and which option was selected the most.
Ordinal scales are used to depict the order of values. For this scale, there's quantitative value because one rank is higher than another.
An example of an ordinal scale is, "Rank the reasons for using your laptop." A reason like job-related functions might rank higher than entertainment.
You can analyze both mode and median from this type of scale, and ordinal scales can be analyzed through cross-tabulation analysis. We'll explain this method later on, but a cross-tabulation analysis compares two sets of data within one chart, like in the example below.
Interval scales depict both the order and difference between values. These scales have quantitative value because data intervals remain equivalent along the scale, but there's no true zero point. This means participants have to record an answer that falls somewhere along the scale.
An example of an interval scale is in an IQ test. The difference between an IQ of 90 and 100 is the same as 100 and 110. However, the intellectual characteristics found at each interval differs as the value increases. Additionally, it's impossible to score a zero on an IQ test as the minimum score is 40.
You can analyze mode, median and mean from this type of scale and analyze the data through ANOVA, t-tests, and correlation analyses. ANOVA tests the significance of survey results, while t-tests and correlation analyses determine if datasets are related.
Ratio scales depict the order and difference between values, but unlike interval scales, they have a true zero point. With ratio scales, there's quantitative value because the absence of an attribute can still provide information.
For example, a ratio scale could be, "Select the average amount of money you spend online shopping." Choices like $75-$100 will rank higher than $50-$75 and the difference between intervals remain the same. But, there's a true zero point since someone may spend $0 on online shopping. Even though this person's answer is zero that response still provides insight into your customer base.
You can analyze mode, median and mean with this type of scale and ratio scales can be analyzed through t-tests, ANOVA, and correlation analyses as well.
2. Select your research question(s).
Once you understand how survey questions are analyzed, you should highlight the overarching research question(s) that you're trying to solve. Perhaps, it's "How do respondents rate our brand?"
Then, look at survey questions that answer this research question, such as "How likely are you to recommend our brand to others?" Segmenting your survey questions will isolate data that's relevant to your goals.
3. Analyze quantitative data first.
Quantitative data is valuable because it uses statistics to draw conclusions. While qualitative data can bring more interesting insights about a topic, this information is subjective making it harder to analyze. Quantitative data, however, comes from close-ended questions which can be converted into a numeric value. Once data is quantified, it's much easier to compare results and identify trends in customer behavior.
It's best to start with quantitative data when performing a survey analysis. That's because quantitative data can help you better understand your qualitative data. For example, if 60% of customers say they're unhappy with your product, you can focus your attention on negative reviews about user experience. This can help you identify roadblocks in the customer journey and correct pain points that are causing churn.
4. Use cross-tabulation to better understand your target audience.
If you analyze all of your responses in one group, it isn't entirely effective for gaining accurate information. Respondents who aren't your ideal customers can overrun your data and skew survey results. Instead, if segment responses using cross-tabulation, you can analyze how your target audience responded to your questions.
Cross-tabulation records the relationships between variables. This reveals specific insights based on your participants' responses to different questions. For example, you may be curious about customer advocacy, but you know your customers are based out of Boston, MA. You can use cross-tabulation to see how many respondents said they were from Boston and said they would recommend your brand.
By pulling multiple variables into one chart, we can narrow down survey results to a specific group of responses. That way, you know your data is only considering your target audience.
5. Understand the statistical significance.
As we mentioned in the last section, not all data is as reliable as you may hope. Everything is relative, and it's important to be sure that your respondents are an accurate representation of your target audience.
For instance, your data states that 33% of respondents would recommend your brand to others. 75% of them were over 40 years old, yet your target audience is 18 to 29 years old. In this case, this data isn't statistically significant as the people who took your survey don't represent your ideal consumer.
Random sampling -- selecting an arbitrary group of individuals from a larger population -- can help create a more diverse sample of survey responses. Additionally, the more people you survey, the more accurate the results will be.
When you run an analysis on software like SPSS -- as shown above -- it will tell you if a data point is statistically significant. If you look just below the table, it says "*. Correlation is significant at the 0.05 level (2-tailed). **. Correlation is significant at the 0.01 level (2-tailed)." This indicates which values are statistically relevant to your business.
If the statistical significance for a data point is equal to or lower than 0.05, it has moderate statistical significance since the probability for error is less than 5%. If the statistical significance is lower than 0.01, that means it has high statistical significance because the probability for error is less than 1%.
6. Take into consideration causation versus correlation.
Another important aspect of survey analysis is knowing the conclusions you're drawing are accurate. For instance, let's say we observed a correlation between ice cream sales and car thefts in Boston. Over a month, as ice cream sales increased so did reports of stolen cars. While this data may suggest a link between these variables, we know that there's -- probably -- no relationship.
Just because the two are correlated doesn't mean one causes the other. In cases like these, there's typically a third variable -- the independent variable -- that influences the two dependent variables. In this case, it's temperature. As the temperature increases, more people buy ice cream. Additionally, more people leave their homes and go out, which leads to more opportunities for crime.
While this is an extreme example, you never want to draw a conclusion that's inaccurate or insufficient. Analyze all sides of the story before assuming you know what drives a customer to think, feel, or act in a certain way.
7. Compare data with that of past data.
While current data is good for keeping you updated, it should be compared to data you've collected in the past. If you know 33% of respondents said they would recommend your brand, is that better or worse than last year? How about last quarter?
If this is your first year analyzing data, make these results the benchmark for your next analysis. Compare future results to this record and track changes over quarters, months, years, or whatever interval you prefer. You can even track data for specific subgroups to see if their experiences improve with your initiatives.
Now that you've gathered and analyzed all of your data, the next step is to share it with coworkers, customers, and other stakeholders. However, presentation is key in helping others understand the insights you're trying to explain.
The next section will explain how to present your survey results and share important customer data with the rest of your organization.
How to Present Survey Results
1. Use a graph or chart.
Graphs and charts are a visually-appealing way to share data. Not only are the colors and patterns easy on the eyes, but data is often easier to understand when shared through a visual medium. However, it's important to choose a graph that highlights your results in an uncomplicated way.
This is an example of a stacked bar graph my team created using data on the brand Allbirds. If you're having trouble reading it, that's okay! We received feedback that it was confusing to understand. That's because the data wasn't organized in a way that would make sense to a stakeholder who's unfamiliar with our project.
So, we decided to revamp our graph's image and came up with the design below.
This bar graph is much simpler to read because it has individual bars for each variable and a clear key. And, the design fits the data that we're trying to display. A reader can easily understand the information we obtained from our survey.
Depending on the survey you've conducted, there are many types of graphs and charts you can use. A few options you can choose from are pie charts, Venn diagrams, line graphs, scatter plots, histograms, pictograms, and more. Be sure to pick one that accurately displays your data and is clear to your stakeholders.
2. Create a data table.
Tables are a great way to share numerical data. You can use software like SPSS or Excel to easily display data, like in the example below.
This table was created from a cross-tabulation analysis. We removed the unnecessary information -- statistical significance, mean, median, etc. -- and focused on the data we wanted to share: the percentage of each gender that preferred each promotional incentive. This gave us a format that clearly demonstrated the percentages we were looking to share with our stakeholders.
3. Make a visual presentation.
Sometimes combining visuals with text creates a thorough description of your findings. In these cases, a presentation could be a good fit for showcasing your data. This gives you a chance to present the earlier stages of your survey, including research questions, hypotheses, survey questions, and methods of analysis.
This slide from my presentation combines a graph with a table and some text. The same data is shared in three different ways to appeal to people with different learning styles: those who prefer visuals, those who prefer numbers, and those who prefer words.
4. Put together an infographic.
If you need to share data that's easy-to-read and quickly consumed, infographics might be your best bet.
This HubSpot Research infographic explains survey results through icons, numbers, and descriptive text. Infographics are incredibly effective for this purpose, breaking down complex ideas into simple messages that are more appealing to read than blocks of text.
5. Present a report.
Sometimes those blocks of text are essential for persuading stakeholders. If you're presenting data to senior executives or business clients, you might want to prepare a full report on your findings. You wouldn't refer to this document during a presentation, but you might hand this to your audience to read through on their own time.
This is the table of contents page from my report on our survey project. It's important to keep track of all the work you've done and maintain records of how you conducted your survey. That way, you won't make similar errors or have to duplicate any research.
If you're not getting the data you're looking for from your surveys, learn how to create great survey questions.
Originally published Aug 9, 2019 8:00:00 AM, updated December 02 2020