Data upload

Zilliant Gen AI Analytics requires properly-formatted transactional data to produce analytics. You must upload data to File Upload in CSV format, where the system validates your files for common errors and accepts or rejects the files. Analytics sends you an email notification if it rejects a file and provides a summary of why.

Before you continue

Analytics validates files only for common file errors such as incorrect line endings. For Analytics to generate meaningful pricing analytics, your files must contain relevant column headers and fields before you upload them to Zilliant.

Your files must also meet Zilliant’s data input specifications. For more details, read Data model.

For best data formatting practices, read Best practices. If you experience issues managing uploaded data, contact your Zilliant representative.

Upload data

1. Select File Upload.

2. Select a storage bucket.

3. If you want to create a folder, select Options, then Create folder. Enter a folder name, then select Create folder.

4. Select Options, then Upload.

5. Select CSV files to upload, then select Start.

6. Track your files’ status and upload progress under the Status and Progress columns. To view your uploaded files, select Back.

View uploaded data

To view the latest version of your storage bucket, select Refresh.

Tip

After validation, a file may take 1–2 minutes to appear in your storage bucket.

To search for an uploaded file:

1. Select File Upload.

2. Select a storage bucket. If you want to search for an uploaded file in a folder, select the folder.

3. Go to the Search current folder search bar and enter a file name. Select the Include subfolders checkbox to extend your search to files in child folders.

Best practices

The system validates files for common file errors. To ensure a valid data upload, check your files against the following best formatting practices.

Files

Your files should:

• Tie out corresponding data sets with other files

• Not contain duplicate data

• Comply with RFC 4180

• Use only lowercase letters, numbers, underscores, and the CSV file extension for the file name

Tip

For consistent file names, we suggest using a <data_model_name>.csv file naming convention where <data_model_name> is the file’s data model. For example, customers.csv.

There can be many causes of duplicative or near-duplicative data, including combining data from multiple sources. To avoid this, make sure to implement data quality processes, such as standardizing data collection and entry procedures.

For more details about RFC 4180, read RFC 4180 .

Columns

Each of your files should have a column header row.

A file’s column headers should:

• Meet data specification requirements as specified in the Analytics data model

• Use only lowercase letters, numbers, and underscores

For example

If a file’s column headers include AccountId, AccountNAME, and Customer_Id, then revise the file’s column headers to all use lowercase letters and underscores. For example, account_id, account_name, and customer_id.

Data rows

Each of your files should:

• Have at least one data row

• Use commas to separate fields

• End lines with CRLF

• Use the ISO 8601 date format for dates

• Not use ASCII characters in numerical data

• Enclose a field that has special characters with quotation marks

The files’ data types must also meet data specification requirements as specified in the Analytics data model. Data rows should list data in the same order as their file’s column headers.

For example

If a file’s column header row is transaction_id, account_id, total, order_datetime, then its data rows should list data in the same order.