Begin by exporting the data from your Postgres database into a CSV or JSON format. This can be achieved using the `COPY` command. For example, to export a table named `users` to CSV, use the following command:
```
COPY users TO '/path/to/exported_file.csv' DELIMITER ',' CSV HEADER;
```
Ensure that the file is saved in a directory accessible by your system.

Open the exported file and review the data to ensure it is complete and accurate. Clean any inconsistent data, handle null values appropriately, and ensure that the data types match the target Convex requirements.

Before importing data, set up the schema in Convex to match the structure of your Postgres data. Define the necessary tables and fields in Convex, ensuring that data types and constraints are compatible.

Create a script in a language of your choice (e.g., Python, JavaScript) to read the exported file and transform it into a format suitable for Convex. This script should handle data type conversions and ensure the data aligns with the Convex schema.

Utilize Convex's API or SDK to establish a connection from your script to your Convex database. This may involve authentication steps, such as API keys or tokens, as specified in Convex documentation.

Use the previously written script to iterate over the cleaned and transformed data, inserting it into Convex. This step involves writing code to loop through the records and make API calls or SDK function calls to insert each record into the appropriate Convex table.

Once the data has been imported, verify its integrity by comparing a sample of records between Postgres and Convex. Run queries to check for completeness, data consistency, and correctness. Make adjustments if any discrepancies are found during the verification process.

By following these steps, you can manually move data from Postgres to Convex without relying on third-party connectors or integrations, ensuring a controlled and customized data migration process.