There are two major components to building FAQs with Flow Builder
- A 'sheet' containing the questions, answers, and associated keywords. This is the knowledge base discussed in the previous section
- An activity containing all of the steps, actions, and transitions we need to create the user experience
The FAQ activity depends on having your questions and answers stored in a knowledge base. This base is created with a CSV sheet. To create a new sheet:
- Navigate to the sheets screen in Studio selecting FAQ's from the flows builder top menu.
- Create a new sheet by selecting the + button next to SHEET ID.
A new sheet will be displayed on the main screen.
- Rename the sheet to something that let you know what it's for.
There are two ways to populate the sheet - manually in Flow Builder or by uploading a properly formatted CSV file.
The typical FAQ activity is limited to 40 questions and answers and is based on semantic search. if you have more than 40 FAQs or would like more advanced modeling. Please contact your Yalo representative.
Manually Adding FAQs
When manually adding FAQs to your sheet, it is as simple as clicking the add row button and filling out the required fields.
The fields are:
- Question - what question will be answered in this row
- Answer - the answer to the question
- Keywords - terms that users may use that should direct them to this answer but the semantic search isn't catching
Keep Keywords to a Minimum
Be careful of using too many keywords. This can cause the semantic search to match an FAQ to a user question when it shouldn't. Keywords should be targeted, and the result of careful analysis should be used to determine where the semantic search is not adequately understanding user questions.
Importe FAQs with a CSV file
The first step is to download a CSV file by clicking the 'Download Data' button. This works equally well for new Sheets as well as existing ones. It ensures that you have a properly formatted file.
Once downloaded, you can open the file in a spreadsheet program like Microsoft Excel or your favorite text editor.
The first thing to note is that the first row (or line) of the CSV contains the field names. The first row of your CSV should always contain the field names and not a question/answer/keyword list. You can use custom field names but these will need to be mapped during the import process
The second thing to note is that there are four (4) fields in the CSV file. In addition to the three discussed previously (Question, Answer, Keywords) there is also
documentId. The document id is assigned by Yalo to each row in your FAQ sheet. It is a unique identifier for that FAQ and is for, among other things, enabling reporting on the performance of your FAQs.
- While you can use your own IDs for this field, it is recommended to let Yalo assign them automatically to prevent duplicate IDs and ensure functionality like reporting is enabled.
- Do not change the
documentIdfor existing FAQs as this will break reporting
- Update the CSV with your FAQs
Each row in your CSV represents a single FAQ. So, if you have 40 FAQs your CSV will have 41 rows -- the header row plus a row for each FAQ.
While the Question and Answer fields are free text, the Keyword field should be a comma-separated list of terms surrounded by quotation marks.
"keyword1, keyword2, keyword3"
- Uploading the CSV file.
Once you have updated your CSV and reviewed it, it's time to upload it via "Upload CSV" button. First, tell Flow Builder what to do with the file.
- Are you adding new FAQs to the existing sheet?
- Are you creating a new Sheet?
- Are you replacing the existing Sheet with new values?
When uploaded, Flow Builder will attempt to validate your file.
When the file is validated, you will need to map the fields in your CSV to the expected fields in Flow Builder
Unless you have a very specific reason to do so, always select "Current ID"
Once uploaded into Flow Builder, your sheet should update with the new FAQs.
Updated over 1 year ago