Analyzing answers is the key to making sense of customer feedback. After all, you won’t benefit from creating even a perfect survey with state-of-art questions and design if you don’t take your time to analyze responses and see what people are actually saying. We observe this mistake too frequently – people create a survey (or several surveys), publish them, and then don’t come back to analyze results. Don’t go this way. In this article, you’ll see how to analyze survey results to make the most of your surveys and feedback you collect.
Establish a routine of analyzing results
Analyzing answers provides the best results when it’s a part of your routine. Otherwise, you can simply forget about your surveys. Recommended interval – daily when you’re just getting started with collecting feedback or create a new survey, then weekly. For example, I check results of my surveys each Monday. Checking answers more frequently at the beginning allows you to check whether people understand the surveys. Then proceed to check a bit less frequently to see the big picture.
In order not to forget about analyzing results, you can use reports (offered by most of customer feedback tools, including Survicate) delivered straight to your inbox with a snapshot of results in a chosen interval.
Use excel to analyze big chunks of data
Browsers are not meant for analyzing data – excel or similar tools are. So if you have thousands of answers to analyze, it’s easier to export them as .csv or .xls file (.xls is often easier to manipulate, especially when you’re not an excel pro) and open it in excel or google sheets than to browse them in an analytical dashboard of a survey tool. It becomes crucial when you use text fields because they produce large amounts of data. If you use radio boxes or checkboxes you’ll see a graph like this so you’ll quickly realize which answers are the most popular.
But if you use text fields in your survey, you’ll see just a long list of answers – not so easy to get through them.
In Survicate dashboard, you can filter responses based on date, browsers, system, and tags (more about them in a second). While analyzing answers in excel, you can filter responses that eg. contain a certain word or phrase, or submitted on a certain page. Using such filters, you’ll quickly make sense of answers, even thousands of them. If you’re not sure how to use filters in excel, take a look at this article: filtering in excel.
Clone surveys to analyze trends easily
This best practice of analyzing survey results lies in between creating surveys and analyzing results. If you want to easily spot trends in answers, it’s the best to clone surveys and run a new one each week or month. Why? Because then I easily see trends. When I analyze one survey that has been running for a few months, I only see a breakdown of answers with no segmentation by time. Of course, I can choose time range but when I run a new survey each month, I clearly see on my dashboard how answers are changing over time and whether they’re going in the right direction.
It’s something for more advanced users. One of the coolest features of Survicate, offered by a handful of other providers, is tagging. Long story short, you can tag people who give a certain answer. Then, you can target another survey at people who have or who don’t have this tag (here’s more about it: Progressive profiling explained).
Sample use-case: run an NPS survey with targeted website survey and tag people as promoters, neutrals, and detractors. Then, run another survey and see how different groups of visitors respond in a month or a quarter.
Track completion rates
Completion rates are one of the most important success metrics of surveys. You don’t want people to just start answering your surveys, you want them to actually complete the surveys. High drop off rates can suggest several problems, for example too complicated questions, lack of relevancy (questions like ‘Please rate your order, if you made any’ in surveys distributed to new customers are still commonplace, unfortunately), or maybe you expect some information participants find confidential.
Look for bottlenecks and analyze what might be causing them, sometimes even changing the wording of a question can decrease drop off rates by a high margin.
It’s easy to track drop off rates in Survicate as we record answers to every single question so you can see which question leads participants to abandon completing the survey.
These are general rules that apply to all kinds of surveys. However, different types of surveys can gather different additional data. So here’s what to look for and a few hints you can find useful while analyzing results to website surveys and Net Promoter Score or in-message surveys.
Analyzing results of targeted website surveys
- take a closer look at browsers and system when people report technical problems.
- analyze where exactly feedback was submitted
When you run e.g. an exit intent survey on checkout pages and see that people point out bugs, see what browsers they are using. You can be surprised how your website behaves on certain versions of browsers – some elements might not load or look completely different than on other browsers.
If you’re running targeted website surveys on multiple pages (as it’s usually the case), look closely on which pages visitors respond to surveys. If you’re using Survicate, download a .csv or .xls file to get this data. Don’t panic when you see several answers pointing out the same technical issue, it does not necessarily mean that your whole website is broken. It can turn out the issue appears only on a certain page and you can fix it in no time.
Analyzing results of Net Promoter Score and in-message surveys
- analyze results in third-party tools, especially marketing automation platforms
You’ll see that Mr. Braun who bought 2 pairs of running shoes, gave you NPS 5 with a comment ‘Too slow delivery’. It’s much more actionable than just analyzing NPS within the analytical dashboard or excel file. You can write a personal email to express your sorry and check details of this transaction. Did it take you too long to send a parcel of maybe the courier company is to blame here? If you want to see whether Survicate works with mailing, CRM or marketing automation tool you use, take a look at our integration directory.
Of course, you also see all those data (including email addresses) when you download survey results as .csv or .xls files.
As you see, analyzing survey results is not rocket science. Remember about doing it on a regular basis (set up reports to make it easier). Over time, when you’ll get used to that and will want to dig deeper in answers, download them as .csv or .xls files. Several simple filters will help you get exactly what you need, no matter how many answers you collect. If you want to see how answers to surveys change over time, regularly clone surveys. And if you’re familiar with more advanced features of Survicate, keep a close eye on tags and send answers to integrated tools.