All Collections
Taxonomy
Managing Keywords and Reasons
Managing Keywords and Reasons

Details on adding, editing, archiving, and merging Keywords and Reasons

Team Enterpret avatar
Written by Team Enterpret
Updated over a week ago

Enterpret's Taxonomy comprises of Tracked Keywords and Reasons, representing "what" the feedback is about and "why" the customer shared the feedback. Here's how you can add, edit, archive, and merge Keywords and Reasons in your Enterpret Taxonomy to further improve it.

Please note that any changes made to your Taxonomy may take up to 60 minutes to take effect and you can check their status on the Taxonomy page. You will receive a notification via email once the changes are completed.

Keywords

Adding a Tracked Keyword

  1. Click on the Taxonomy icon from the left nav bar to navigate to the Taxonomy section

  2. Click on the "+ Tracked Keyword" button in the top right corner (if you don't see it, make sure you are on the Tracked Keywords tab)

  3. Fill in the required information:

    1. Name: This is how the keyword will display in Enterpret

    2. Phrases: These are the constituent words or phrases that define the keyword. When adding phrases, you don't need to add plurals or syntactical variations (like hyphenation). In many cases, you will only need a single phrase, and it will be the same as the keyword itself. In other cases, you may decide that it's appropriate to have additional phrase representing synonyms. Here are some real customer examples:

      1. Download keyword that includes the word itself and one additional phrase

      2. Performance keyword that includes a wide range of phrases to capture the various ways that customers might discuss the broad category of performance issues – including "fan", in case they talk about the loudly whirring fans on their laptop!

    3. Duration: How long back Enterpret should look when applying this keyword to feedback records. Adjust this according to your needs - e.g., if you just released a named feature a week ago and want to track references to it, a 1 Month lookback will suffice.

    4. Description: This field helps provide context to our models by flagging words that refer to specific product features or competitors. For any keywords that don't fit in either of those categories, just leave this as "None". For example, if you were AirBnb, you might have the following keywords:

      1. Filters: Feature

      2. VRBO: Competitor

      3. Host: None

  4. Click on "Add Tracked Keyword" in the top right corner to finalize the keyword

Editing a Tracked Keyword

  1. Navigate to the Taxonomy page

  2. Hover over the row containing Tracked Keyword you want to edit, and click the "Edit" icon all the way to the right

  3. As needed, make changes to the keyword name or phrases (refer to the Adding a Tracked Keyword section above for more details on phrases)

  4. If you modified the phrases, select a Duration to tell Enterpret how far back to look in attempting to re-tag based on phrase changes. In most cases, "All Time" will make the most sense in order to ensure consistency.

    1. A name change is always applied to the full history of the keyword, and if you only changed the name, then the Duration field does not matter here

  5. Click on the "Save" button to save the changes

Archiving a Tracked Keyword

Archiving a Tracked Keyword will remove it from all feedback records in your instance. You should archive a keyword if you feel that it is not providing value to your team, e.g., it is too general or common to be useful.

  1. Navigate to the Taxonomy page.

  2. Click on the "Archive" icon next to the Tracked Keyword you want to archive.

  3. Confirm by clicking on the Archive button.

  4. You can always restore an archived keyword by reaching out to our team here

Reasons

Merging Reasons

While analyzing your data, you might come across similar reasons that you'd like to merge for a more streamlined analysis. When you merge two or more reasons, all feedback records associated with any of them will now be associated with the new reason.

Some things to know about merging reasons

  1. Merging reasons will apply for all users in the instance, not just you

  2. Merging reasons can be undone, so don't hold back on merging if it will improve the quality of the instance... but we'll have to help you with the un-merging. Reach out to your customer success manager or system admin, or use the contact form here

  3. After you initiate a reason merge, it may take up to 60 minutes for your changes to become visible. Both the new and the original reasons will disappear from any charts and analyses during this time, so don't be alarmed!

  4. Keep an eye on the Auto-namer! When you are merging reasons, Enterpret will suggest a new name for the combined reason based on its components. Usually this Auto-namer does a good job, but always check the final name before merging. Sometimes users don't notice that the Auto-namer has made a change and then struggle to find their new reason later on!

Where to Start a Reason Merge

You can start a Reason merge from a few different places:

  1. Taxonomy page

    1. Merge Suggestions: Click this button at the top of the Taxonomy page to see suggestions for potential merges. You can filter by category, or enter search terms in order to focus your efforts. If a search suggestion isn't helpful, feel free to press the "Dismiss Suggestion" button to hide it.

    2. Manually select Reason(s) to merge: Scroll or search in the Taxonomy page and then use the checkboxes to select Reasons to use as a starting point for a merge, then press the "Merge Reasons" button in the upper right. You will have a chance to add more, so don't worry about trying to get them all here, and you can even start with just one.

  2. Quantify page

    1. From the chart: Hover over one of the bars (bar chart mode), individual data points (trend mode), or Reason names in the legend (any chart mode) and click "Merge with another Reason"

    2. From the table: Click on the 3-dot menu on the relevant row in the table beneath the chart, then click "Merge with another reason"

The Merge Reasons Flow

Wherever you started, you'll eventually end up in the Merge Reasons flow. In this flow, you select reasons to be merged, confirm the category and name of the new reason you'll create, and then save.

  1. Category and Name of merged reason

    1. The Auto-namer will update these based on the reasons you add, so always check them – especially the name – before saving!

  2. List of reasons that will be included in the merge, with feedback count

  3. Search bar to search for additional reasons to add

  4. Category filter on search pane

  5. Search results, or suggested reasons based on similarity to your selection

  6. Buttons to add a reason to the merge (+) or remove a selected reason from the merge (✓)

  7. Button to regenerate the suggested reasons – try this after adding a few reasons

  8. Save button to commit the merge

Creating Reasons

While analysing the reasons created and provided by Enterpret, you might feel the need to create & add new reasons. There are three 'Create Reason' journey entry points, which will be covered here.


Journey 1 : Through the Taxonomy Page

Here are the steps that you'll need to follow :

1. Navigate to the Taxonomy page.
​


2. Click on the ADD NEW dropdown button on the top right of the screen, Click on Add new Reason from the dropdown.
​


3. A new pop-up window will open up.


Step 1 of 2 :

You'll have to describe a broad idea about what kind of feedback do you expect the reason to predicted on.
​
Then click NEXT at the bottom right of the pop-up window
​
Let us take an example here.
​


On clicking NEXT, there would be a loading state visible in which Enterpret fetches and curates definitions for the reason.

Definitions are the building blocks for Reasons. They are repeatable ways in which users talk about problems.


4. After Enterpret is done fetching the Definitions, you'll be shown some selected Definitions which will be used to find relevant feedback.

Step 2 of 2 :

Check if they fit in what you want to talk about.
​


Also, you'll be provided with a search bar to Find and add more Definitions at the bottom right half of the pop-up window.
​


View the feedback associated with the Definition and check if it is relevant to your reason.
​


The Feedback record with its summary will be opened in a new browser tab. (In this case, it is not the right match for what we are talking about, hence we'll not add this to the Definitions)
​


Now coming back to the Pop-up window, you are provided the Category and Reason Name generated by Enterpret. You can regenerate both of them or else edit them manually.
​


To complete this step and move forward, click on CREATE REASON at the bottommost right of the pop-up window.


A new reason will get created and a small pop-up window will appear on your screen. Kudos! you have created a new reason for the first time 🎉

Journey 2 : Through any Feedback record

The second user journey through which you can create reasons is through the feedback records.
​
For example let us go through the search page.
​


Go to a particular feedback record, click on the plus icon beside Predictions, then, Click on Create new Reason from the expanded panel.
​


​Follow the same steps onward as were followed in the Journey 1 after the Pop-up window appears.

Journey 3 : From another Reason

There is another entry point for creating reasons, through Another reason. The steps for this journey are :

  1. Navigate to the Taxonomy page > Reasons.
    ​

  2. Click the pencil icon (Edit) on any of the reasons' row.
    ​

  3. A hovering Pop-up window will open up on the right side of the screen.
    ​

  4. You can now see a list of Definitions in a box at the bottom right.
    ​
    Let's take an example : For this Reason 'Happy with The User-Friendly UI', you feel that 'happy with the reliable performance of zoom' Definition is something that should be tracked separately for some reason.
    ​


    Click on the Double Comma Icon : Breakout into a New reason with this Definition.


    The pop-up window to create a new reason will appear, Follow the same steps onward as were followed in the Journey 1 .
    ​
    NOTE : Once, a new Reason is created from this Definition, that particular Definition will now not be shown in the previous reason anymore, as it has now been separated out.
    ​
    In this example, now, after creation of a new reason, the Definition 'happy with the reliable performance of zoom' will no longer show up in the list of Definitions of the Reason 'Happy with The User-Friendly UI'.

Archiving Reasons

Archiving a Reason will remove it from all feedback records in your instance. You should archive a Reason if you feel that it is not providing value to your team, e.g., it is non-sensical or is too specific to be broadly meaningful. Sorry, sometimes the models don't behave :/

  1. Navigate to the Taxonomy page.

  2. Click on the "Archive" icon next to the Reason you want to archive.

  3. Confirm by clicking on the Archive button.

  4. You can always restore an archived Reason by reaching out to our team here

Related Articles
Taxonomy Management
Enterpret Features Explained
Managing Taxonomy in Enterpret?
What are co-occurring predictions?
Did this answer your question?