Begin by logging into your TikTok for Business account. Navigate to the analytics section where you can access data related to your campaigns, including metrics like impressions, clicks, and conversions. Use TikTok’s built-in export functionality to download the data in a CSV format. This file will serve as the raw data source for your migration process.

Ensure that your local environment is set up for data manipulation. Install Python and any necessary libraries such as Pandas, which will help in processing the CSV file. Additionally, ensure DuckDB is installed and accessible from your command line. DuckDB can be downloaded from its official website and installed according to your operating system.

Open your Python environment and use Pandas to read the CSV file. Conduct any necessary data cleaning operations such as removing duplicates, handling missing values, and ensuring data types are consistent. If needed, transform the data to fit the schema you plan to use in DuckDB. This might involve renaming columns or changing data formats.

Open a terminal or command line interface and start a DuckDB session by typing `duckdb` to enter the DuckDB shell. Create a new database, if one does not exist, by executing `CREATE DATABASE tiktok_marketing;`. This command initializes a new DuckDB database where you will store your TikTok data.

Define the schema that matches the structure of your cleaned and transformed data. In the DuckDB shell, use SQL commands to create a table. For example:
```sql
CREATE TABLE tiktok_data (
campaign_id INTEGER,
impressions BIGINT,
clicks BIGINT,
conversions BIGINT,
spend FLOAT,
start_date DATE,
end_date DATE
);
```
Adjust the column names and data types to match your dataset.

Use DuckDB's shell to load the CSV data directly into the database. In the terminal, execute the following command:
```sql
COPY tiktok_data FROM 'path/to/your/cleaned_data.csv' (AUTO_DETECT TRUE);
```
Replace `'path/to/your/cleaned_data.csv'` with the actual path to your cleaned CSV file. The `AUTO_DETECT TRUE` option allows DuckDB to automatically infer the correct data types for each column based on the CSV content.

After loading the data, run several SQL queries to verify that the data has been imported correctly. Check for the correct number of records, validate data types, and ensure all fields contain the expected values. For example, you might run:
```sql
SELECT COUNT(*) FROM tiktok_data;
SELECT * FROM tiktok_data LIMIT 10;
```
These queries help confirm that the data migration process was successful and your DuckDB instance is ready for further analysis or querying.