All Collections
Results and reports
Analysis of statements (open answers)
How to perform a quantitative analysis of the verbatim? (cross tables, rankings)
How to perform a quantitative analysis of the verbatim? (cross tables, rankings)
Artur Zbiejcik avatar
Written by Artur Zbiejcik
Updated over a week ago

Creating tags and then analysing the feedback from the clients allows for a quantitative analysis of qualitative data, which we understand as comments and emotions of the respondents towards our brand. From the first part of the manual you learned how to create tags - the categories according to which the feedback will be analysed and how to tag the respondents' statements with the emotional color.

It is the time for another portion of detailed, practical knowledge - in this section we explain how to analyse the collected and tagged information in a quantitative manner, thus giving specific data and possible to be presented in tables and graphs. For this purpose, we will use the advanced capabilities of the Startquestion tool, such as:

  • creating cross tables with tags, which allow us to get the information about what and how many different respondents say about the brand,

  • rankings - a list of the most frequently and least discussed topics,

  • creating trends - obtaining information on how the frequency of the topic (tag) in the respondents' comments changes.


Thanks to the analysis of the verbatims with the use of the advanced cross tables, we can create a list of any question in a selected survey addressed to the brand's clients / users with selected tags (for example, price, product, quality). Crossing the tags with the answers from the selected question will allow us to get the information about what the people who marked specific answers were talking about.

When choosing the NPS question for the analysis, we will find out what and how much do the critics, neutral and promoters say. After crossing the tags with an evaluation question, we will learn the same about the satisfied and/or dissatisfied people

Therefore, all these analyses have one common denominator - accessing the information about what and how many different respondents say about the brand.

1. To begin the analysis, select the Results tab, and then Cross tables. To create a new table, just click on the green text: Create a new cross table in the middle of the screen.

2. From the available options when creating a table, first select the appropriate tag or tags previously created in the account, then select the survey question which interests us and click the green button: Create cross table.

As a result, we get a graph and a table containing the information on how many promoters, critics and neutral people spoke about a specific tag, for example, the reasons to quit their job.

The user can change the view of the graph (positions of the XandY axis) and presenting the data - as a percentage or number. The cross table created this way can also be added to the report (the button in the upper right corner of the table).


The main task of the rankings is to rank the tags selected by the user (the issues that the clients talked about in the comments) from the most frequently to the least discussed ones. From the ranking, we will find out how much is said on a given topic and what the NPS or LTR indicator (depending on the question we are analysing) the people who talk about it have.

1. To create a ranking, go to the Results tab, and then Rankings. A button will appear in the middle of the screen: Create a new ranking.

2. Name your ranking, next select the tag or tags to be included in the ranking, and then the survey question to be analysed (for example, the NPS question here).

After creating the ranking, you will see at a glance which topics the clients are talking about the most and the least about. So your company will get a clear list of which topics are more important for the clients and which are not so important for them. The results in the table can be sorted according to the type of the question. In the case of the NPS, it is possible to sort by a given group (critics, promoters, etc.) or LTR or the number of responses, using the arrows next to the column's name.

The ranking created in this way can be additionally filtered or added to the report.


By creating trends, we are able to find out how often a given topic (tag, for example, the price, quality) was mentioned in weeks, months or quarters, and how did it change over these periods. We get a picture of what part of a given tag in what the respondents generally say.

1. The trends can be created by going to the Results tab, and then Trends:

2. In the middle of the screen, the option: Ceate a new trend will appear, after clicking on which the trend wizard will appear, where the user must complete the name and select the trend range (which means, the time in which the analysis is to be presented - month, quarter or week). Then select the tag tab and select a category or several tags which you want to analyse.

3. By clicking Create a new trend, we will receive a graph and a table showing how often the respondents spoke about a given topic during a month, week or quarter.

At any time, you can change the view of the chart from the area line to a bar chart and from a percent to number values. You can also add a chart and table to a report. It is possible to filter the results of the trend by the filter created earlier:

This is all about the quantitative analysis of the client's statements based on the tags. You can find more information on the advanced tools: cross table, rankings or trends in our manual and on the Startquestion's blog. In the next manual, we will explain what a sentiment analysis is.

Did this answer your question?