a) Filtering Options
At the top of the All columns, Results and the Text analysis dashboard too, you can find an ‘Add new filter’ button. Use this if you only want to look at certain sections of your dataset.
You can filter by polarity, any of the columns in your dataset such as time or categories, as well as including or excluding automatically recognized labels, custom labels, and custom label categories. Once you’re happy with the selected filters, click ‘Apply filters’. The whole dashboard will change according to your selection, and the applied filters will be shown at the top. If you don’t need them anymore, you can delete them one by one or clear all of them together. You can also always choose to add more filters.
b)Text Search
The Text Search function allows you to search for specific words or phrases across verbatim responses in your dashboards. You can use it to locate relevant feedback, compare subsets of data, or set up targeted alerts and exports.
Text Search is available among filters and applies to the entire dashboard. You can save it as part of a filter set or use it in comparisons, alerts, or scheduled exports - just like any other filter.
Limitations
- Works on up to 100,000 records on general dashboards.
- Text search may add up to 1 minute to the dashboard load time due to intensive processing.
- Supports multiple verbatim dimensions. For example, you can filter survey responses to the “Why?” question based on whether the “Which products did you like?” answers contain the word “Zurvey”.
Search Operators
By default, Text Search uses “contains” logic. For more precise filtering, additional operators are available:
- Contains – finds all records that include the given word or phrase (e.g., good, not good, good :)).
- Starts with – matches words that begin with the given text (e.g., good → matches "goodness" but not "not good").
- Exact match – matches full words exactly as entered.
Note: “Starts with” and “Exact match” only support words without special characters.
Multiple Search terms
You can add multiple search terms (minimum 3 characters each). Press Enter after each term to add it as a tag.
You can define the logical connection between terms as:
- Any – returns results containing any of the listed words.
Example: contains any of bank, good → finds texts with either or both terms. - All – returns results containing all listed words.
Example: contains all of bank, good → finds texts that include both.
Combining Text Search filters
You can apply multiple Text Search filters on a dashboard, this allows you to combine different text conditions across verbatim dimensions e.g. different verbatim questions in surveys.
When multiple Text Search filters are applied, they are combined using AND logic across components.
Example:
- Survey question_1 “Which products did you like?” contains Zurvey AND Survey question_2 “Why?” contains design
c) Saved Filtersets
As you can see, a set of elaborate filters can be created with the available options. Luckily, you don’t have to recreate them each time you open the dashboard: if you are happy with a set of filtering rules, you can save the filterset for later. Just click the ‘Save Filterset’ button on the right and give the set a name.
Your saved filtersets will appear in a drop-down menu afterward – also at the right side of the filtering panel –, and you can simply click on any of them to apply them to your dataset.
d) Comparison Mode
If your datasets are huge, you might need to divide them into data series to make comparisons between them. This is what Zurvey.io’s Comparison mode allows you to do. At the top of the left panel of the dashboard, click on Comparison and then on ‘Add new data series’. Here you can define a data series using your current filter set, previous filter sets or create a new one from scratch using the usual filter selectors (see filtering options above).
Give the data series a name and save it - the color can be modified as well to make it on-brand or for better differentiation. When you have defined the first data series, it appears in the left column list beside an automatically created ‘Overall (No filters)’ one.
Almost every chart has a comparison version on the dashboard which will now appear – there are only a few exceptions such as the Mention graph or the Key driver chart, which cannot be visualized this way.
Any number of data series can be created, but no more than three can be active and visualized at the same time. To switch between data series, you can activate and deactivate them using the checkbox on the data series element. Like filter sets, defined data series are saved in the group too, so you can use them later.
When you’re in Comparison mode, adding filters the usual way is not available, as the data series already show a filtered state.
The comparison feature is available on the ‘All columns’, Results and the ‘Text Analysis’ tab as well. It means both quantitative metadata and qualitative text analysis results are comparable, but it works even if there’s no textual data in your dataset.
Read more about this feature on our blog:
If you eat big datasets for breakfast, this feature is for you