View results
To view the results from your survey, you go either to the overview under “Data sources/Overview”, scroll down and select “Browse results” from the survey module…
…or you go directly to “UX-surveys”.
Select the survey you want to look at in the sidebar menu. As you will notice, there are more options in the sidebar menu after selecting a survey. These are the result headings, or survey question categories, which help you to more easily find the question you want.
There are filter options on the right side, but we’ll get to them in a minute.
Before we look at the chart, if you have your survey in more than one language (or versions, but that’s for later), you can select the displayed version at the top of the page. The displayed version determines which questions will be displayed and in which language.
For example, you may have a survey in Finish but you yourself don’t know any Finish. If you have the same survey in a different language, say English, you can display the survey in English and use the filter to select the Finish results, and then be able to understand the data.
Viewing your data
Before we get to the more complicated things, such as filtering data, let’s start by looking at the charts and see what you can do here.
Single and multiple choice questions
Single and multiple choice questions will be displayed as bar charts, but you are able to change to other chart types if you prefer.
You can choose between percent and number of responses.
You can also switch axis…
…and sort the options. Click again to change the sorting order.
However, only sort the options when you are looking at just 1 segment and 1 time period.
Lastly, you are also able to view the results as raw data table. Here too you can choose between percent and number of responses.
In the bottom right corner there are more options, but not related to viewing the chart. For singel and multiple choice questions, you can cross-tabulate with one of your KPI questions (for example rating question).
Click on the scatter plot icon and select the question you want to compare to. The cross-tabulation will show you the relationship between the two questions. You can read more about cross-tabulations here.
The next icon is for saving the chart as an image, which means it will be downloaded to your computer. This is useful if you want to add the chart to a presentation, for example.
Note that the question is not included in the image, but it will be the name of the image so you’ll still be able to tell the charts apart.
If you don’t want the chart but rather the raw data, you can download as an Excel file instead.
This will give the same information as the chart or the “View raw data” option, and is useful if you want to make your own charts.
Lastly, you have the option of adding the chart to a report, but we won’t go through that here. Instead, you can read more about it in part 7 – Automatic reports.
Text answers
Your text answers are displayed differently from your single and multiple choice questions. First, there’s a word cloud showing you the most common words.
Below is a table with all the answers for the selected time periods.
The default is to show 10 answers per page, but with more than 50 answers, you can select how many to show per page, with up to 100 answers per page.
Categorizing text answers
However, the table is not the easiest way to analyse your text answers, at least not once you have more than 20 answers. Therefore, you also have the option to categorize your answers. Click on the pen and paper icon to open the categorization tool.
Here you’ll be able to see the same information as before, but the table will also contain the columns “Tag” and “Sentiment”.
The tags are your categories, which you choose and create yourself. The sentiments, on the other hand, are always the same: Positive, Neutral, and Negative.
The sentiments are an easy way to categorize the tone of the answers, so that you later can filter for only the positive comments, or only the negative, for example.
The tags are much more specific that the sentiments, and are categories for the content and the tone. For example, if you have text answers for “Other role”, you may need categories such are “Engineer”, “Technician”, and “Potential customer”.
Here, we’ll only focus on how this tool works, but if you want to read more about which categories to use and how to think while categorizing answers, we have an article for you here.
To add categories, or tags, to your answer, you’ll first need to add some tags.
Two things before:
- Read through the first of your answers to find which tags are needed, and add more as you discover new categories. In other words, don’t predefine your categories, but base them on the answers you have received.
- If you have grouped surveys, for example language versions, you will have the same categories for all the surveys in the same group. That means that you need to select the language you want for you categories and use them for all the survey, regardless of the language of the individual survey.
Alright, to add a tag, click on “Select tag” and write your new tag at the bottom of the list. Press Enter to add the tag. Let's call this one "Example tag".
As you can see, the tag has been added to your list of tags in the table as well. It is now ready for use.
To add a tag to and answer, you simply click on the tag.
You remove a tag the same way, by clicking on it. You can also add more than one tag per answer, but to make the results easier to analyze, we recommend that you try to limit to 1 tag in most cases.
The categories, or tags, that your have added are not final, but you can change them if needed. Select the tag you want to change.
If you have made a spelling error or found a better phrasing, you can give your tag a new name.
However, if you have used the tag before, don’t give it a name that has a different meaning from the previous name, as that will complicate the analysis. If you need a different tag, add a new one instead.
You are also able to remove tags. However, you can only remove tags that have not been used. If you want to remove a tag that has been used, you first need to remove the tag from those answers. To do that, start by selecting a date range that includes all uses of the tag. Then, open the tool for categorizing answers and select the tag.
Now you will be able to see how many answers have this tag…
…and you can also remove the tag from those answer with the use of a button.
Once the tag is not connected to any of the answers, you are able to remove it again.
Apart from the “Remove from X responses”, you also have the “Add to X responses”. This button is useful if your respondents are using searchable key words.
For example, if you have several respondents that answered something related to jobs as their purpose, you can use “job” to filter your responses…
…then select the appropriate tag, in this example “Job opportunity”…
…and then add the tag to all the answers you have filtered out.
Apart from filtering by searching for words, you can also filter by tag and sentiment…
…and you can also sort the answers based on the different columns.
After you have added tags to your text answers, you will get an new chart above the text answers, showing you the categories.
This means that you now have a chart for your text answers as well, and unless you’re very strict with “1 tag maximum per answer”, this will be a multiple choice equivalent. The chart gives you a much better overview of what the respondents said, than simply a list of text answers.
Flagging personal data
Extellio has an automatic system for picking up potential personal information, which is then manually handled. Any personal information such as names, phone numbers, email addresses, etc, is masked and replaced with e.g. . While the answers are being processed, you will not be able to see them. Once we have approved or masked personal data, you will see the answers.
One thing to note is that the automatic system isn't 100% accurate. Occasionally, personal information slips through. If you find any personal information in your data, you need to manually flag the comments. You do so by clicking on the three dots by the answers, and then select "Mark for deletion of personal data".
You are able to flag comments both in the results table and in the categorization tool.
When you refresh the page, the flagged answer will be hidden from you, but you will see it again once we have processed it and masked the personal information.
Filtering your data
Now that you understand the charts, let’s look at the filtering options. Here you have to way of filtering your data: by date and by survey responses/respondent information.
Selecting date range
The default setting when viewing your survey data is to see the last 30 days. However, you have lots of options here.
To change the date range, click on the box with the date.
You will now get several options.
At the top, there are presets to choose from: last week, last month, last quarter, and last year.
These settings are based on the calendar. Last week is the last calendar week, and not the last 7 days. Last month is the last calendar month (if you’re in March, that would be February), and not the last 30 days. Same goes for quarter and year.
However, in most cases, you probably want to configure the date range yourself.
To start with, you chose between 4 options: Last, This, Custom date, and Custom weeks.
Last
For “Last”, you select the time period (day, month, etc) and how many you want.
As for the presets, this is based on the calendar. If in March, “Last 2 months” will be January – February.
If you want the last two months, but not based on the calendar, you can select “Last 60 days” instead.
This
The options for “This” is similar to “Last”, but you can only select one time period. For example, “This month”. If you select “This month”, that will again be based on the calendar. When in March, that will only give you data for March, regardless of when in the month you select this date range.
If you select it on the 5th of March, you will only get 5 days of data. If you select it on the 25th, you will get 25 days of data.
If you wanted this month, but not based on calendar months, you can use the date range “Last 30 days” instead.
Custom dates
The option “Custom dates” is useful when you need a specific date range, or a date range that will remain the same over several days. For example, the settings “Last 30 days” will shift every new day, so if you are extracting charts, you want to make sure the chart have the same dates every time you download them.
To select the dates you want, you can either write the dates…
…or use the calendar to select the dates.
Custom weeks
If you work in weeks, you can also select “Custom weeks”, which works similarly to “Custom dates”.
Write the weeks you want, or use the calendar to pick weeks.
Comparing period
Lastly for the date range, you also have the option to compare your data to a previous time period. Again, there are several options for you.
“Previous period” is the same setting as above, but the period before. For example, if you selected “Last 30 days”, the previous period is another 30 days back. If you selected “This month”, the previous period is the month before.
“Previous year” is exactly the same date range, but the year before. For example, if you are looking at Jan 15 – Feb 14 2024, the previous year will be Jan 15 – Feb 14 2023.
Lastly, you can again choose “Custom dates”. This is mainly useful if you are comparing two similar periods in you data collection, for example campaigns or events, that happened at two different times of the year.
For all comparing periods, you can choose up to 3 periods.
Creating segments
When viewing your data for the first time, you will see all the responses. However, for more in-depth analysis, you probably want to segment your data.
You find you segments to the right, and for now, there is only the segment “All”.
This segment cannot be changed, but it can be turned (toggled) off.
However, to do so, you’ll need to activate another segment first, as you always need to have at least 1 segment active at a time. Since there are no other segments to choose from yet, we have start with creating some.
There are two ways of creating segments. One is more detailed, while the other is simpler. Since the more detailed will explain the segmentation better, we’ll start with that one – but if you want to skip ahead to the simpler version, you can do so.
Creating segments from scratch
Start by selecting the “Create segment” button that is located under your list of segments. Here you have a full list of settings.
Segment name
The only required segment setting is the name. Use a short but descriptive name. This name will help you to select the right segment later on, but it will also be used in the chart legend, which means that it needs to be understandable for those looking at your data as well.
For example, if you create a segment based on the role “Existing customer”, you can call you segment “Existing customers”. But if you create a segment based on the ratings 4 and 5, you should not call your segment “4-5”. Instead, you can call it “Satisfied users”, or “Gave good ratings”, or similar. Basically, the segment name should not require any explanations.
Shared segments
Next up, we have the option to share your segments. Generally, the segments are unique to each user. However, by sharing your segment, all other users in your account will get the same segment.
Only you are able to edit a shared segment you have created, and there are license limitations for how many shared segments you can have.
For shared segments, you can also choose to make them available in your analytics data, but we’ll explain that more in part 6 - Combining modules.
Filter category
Let’s skip some of the settings for now, as they will make more sense later, and look at the filter categories instead.
“Survey” are filter options based on which survey version or language. This is the setting needed for the example in the beginning, where you wanted only the Finish survey. With only one active version, these settings are irrelevant for you.
“Questions” is most likely the category you’ll use most, as these are filters based on survey answers. Here, you find the question you want and select the options you want to use for your filter.
You can select several options, and even several questions. However, when using several questions for your filters, it’s easy to accidentally exclude too many respondents and ending up with useless segments. Therefore, try to use only one or two questions for your segments.
“Tagged comments” work similarly to the question filters, and you select which question and which categories you want.
For example, you may want to know what those that struggled with the navigation (segment based on tagged comments) were struggling to do (what they answered for the purpose question).
“Respondent” is information about the survey respondents, for example device used when answering the survey, or if they signed up for the client panel.
Lastly, “Website interaction”. These filters are based on more respondent information, such as page where they answered the survey. If you also have analytics with Extellio, you’ll have more options here, which is explained more in part 6.
After you have selected your filters, let’s look at the settings we skipped.
If you have based your segment on more than one category or question, you can decide if you want the combination rule to be “And” or “Or”.
For example, you have selected the role “Customers” and the purpose “Looking for product information”. You can choose between “Customers AND Looking for product information”, or “Customers OR Looking for product information”.
The default “And” will mostly likely be the one you’ll use.
You are also able to invert segments.
If you have selected the role “Customers”, inverting the segment will make it “All respondents EXCEPT customers” instead.
With the settings done, click on “Save”, and you have a new segment.
Now that you have more than one segment, you are able to choose which of them you want to look at, or even both.
Creating segments from charts
If creating segments seem complicated, don’t worry. There is a shortcut. Say you want a segment based on your customers. Instead of the “Create segment” button, you can go to the role chart and just click on the bar for customers.
The same segment editor will open, but you will have the name and the segment category filled in for you. You are of course able to change the settings if you want to, but if it looks good to you, you just have to click on “Save” and your segment has been created.
You can create more complicated segments this way too. Say you want to create the segment example from above, which was customers looking for product information. You already have a segment for customers, so with that segment active, go to the purpose chart.
Now, click on the bar for the product information.
Again, the segment editor will open, but as you can see, your segment is now based on two questions: the role “Customer” and the purpose “Looking for product information”, with the rule “And”.
At this point, you probably need to edit the segment name.
Once happy, save and your new segment has been created.
There is a lot you can do and learn while looking at your data. If you feel you need some help with the analysis, we have an article to help you.
If you need more help with analyzing your data, you can purchase analysis sessions or analysis reports from our skilled team. Find the product for you here.
Speaking of analysis, in the next part, we’ll look at how you can combine Extellio’s different modules, and what that will bring to your data analysis. If you only have one module right now, you can skip ahead to the automatic reports instead.
For more analysis, let’s move on to the next part.
