Begin by exporting the data you need from Harvest. Log in to your Harvest account, navigate to the Reports section, and select the data you wish to export, such as time entries, invoices, or projects. Harvest allows you to export data in CSV format. Save this CSV file to your local machine.

Open the exported CSV file using a spreadsheet application like Microsoft Excel or Google Sheets. Review the data to ensure it is correctly formatted and contains the necessary fields. Make any required adjustments, such as renaming columns or removing unnecessary fields, to align with your MongoDB database schema.

Ensure MongoDB is installed on your system. If not, download and install it from the official MongoDB website. Additionally, install MongoDB tools like `mongoimport`, which will be used to import data into your MongoDB database. Verify the installation by running `mongo --version` and `mongoimport --version` in your command-line interface.

Open your MongoDB shell or a MongoDB client and create a new database and collection where you will store the Harvest data. For example, you can use the following commands in the MongoDB shell:
```shell
use harvest_db
db.createCollection("harvest_data")
```

Use a script or a command-line tool to convert your CSV file to JSON format, as MongoDB requires data to be in JSON format for import. You can use Python with libraries like `pandas` to read the CSV and convert it to JSON:
```python
import pandas as pd

# Read CSV file
csv_file = 'path_to_your_file.csv'
data = pd.read_csv(csv_file)

# Convert to JSON
json_file = 'path_to_your_file.json'
data.to_json(json_file, orient='records', lines=True)
```

Use the `mongoimport` tool to import the JSON data into your MongoDB collection. Run the following command in your terminal, replacing placeholders with your actual database, collection names, and file paths:
```shell
mongoimport --db harvest_db --collection harvest_data --file path_to_your_file.json --jsonArray
```

After importing, verify that the data has been correctly imported into MongoDB. Use the MongoDB shell or a client to query the collection and check a few entries to ensure data integrity. For example:
```shell
use harvest_db
db.harvest_data.find().pretty()
```

By following these steps, you can successfully transfer data from Harvest to MongoDB without using any third-party connectors or integrations.