Understanding The Data File Columns
All products (goods and services) that are available in your digital assistant are organized in a table view for easy management.
Each row represents a product and each column defines an attribute (product detail) such as name, price, size, color.
Whenever you upload a data file, zoovu automatically structures and categorizes product attributes using predefined column types:
- BOOLEAN — The column value has one of two possible values, usually ‘true’ and ‘false’ or ‘Yes’ and ‘No’.
- TEXT — The column value contains letters and words.
- DECIMAL — The cell value contains decimal numbers
- INTEGER — The column value contains whole numbers that do not have a fractional portion in it.
The column types are assigned automatically when uploading a data file. This can be changed if you want to tweak the results of your digital assistant.
Mandatory Attributes (Columns)
Zoovu defines 5 mandatory columns that must have a value:
- Offer URL
This is a bare minimum to set up your fully functional digital assistant.
The name of the column is not relevant, you can set it manually during importing products, e.g. name vs product_name. It can be adjusted to display Name to the user, instead of product_name as it appears in the data.
A digital assistant can only be as good as its underlying data. If you want to make your technical attributes easier to understand for customers – Computed Attributes are the way to go!
This option allows you to create attributes, whose values are automatically connected to other product parameters. This enables you to add the correct results mapping for your need-oriented questions.
Learn how to easily set up Custom Attributes here or let us know at email@example.com for a personal consultation.
A Quick Example
Let’s imagine you’ve created a bike assistant and you want to ask your customers if they need a product that is easy to be transported. You are going to base that question on the size of the frame and wheel diameter.
Now the top matching product is presented to the customer like so:
The attributes marked green do not clearly indicate if the bike meets the customer requirements. Your target Product Attributes would rather be: