How to Import CSV in pgAdmin

Before you start: pgAdmin import CSV checklist

The Import/Export Data dialog copies data from a file into a table (or from a table into a file). It does not create the table, so you need one whose columns match the CSV. Save the sample below as customers.csv in UTF-8, and create the table in the Query Tool.

full_name,email,signup_date
Ana Silva,ana@example.com,2026-01-15
Ben Okafor,ben@example.com,2026-02-03
"Chen, Li",li.chen@example.com,2026-03-21
Dara Murphy,,2026-04-02

CREATE TABLE customers (
    customer_id integer GENERATED ALWAYS AS IDENTITY PRIMARY KEY,
    full_name   text NOT NULL,
    email       text,
    signup_date date
);

Import a CSV file into a table, step by step

1. In the Object Explorer, expand the database and schema and select the customers table.
2. Choose Tools > Import/Export Data....
3. On the General tab, move the Import/Export switch to Import (the default).
4. In Filename, type the path or click the browse icon (ellipsis) to pick customers.csv.
5. Set Format to csv and Encoding to the file's encoding, for example UTF8.
6. On the Options tab, move Header to Yes so the first row is treated as column names, and check that Delimiter is a comma and Quote is a double quote.
7. On the Columns tab, remove customer_id from Columns to export/import, because the CSV has no value for it and the identity column fills it.
8. Click OK. pgAdmin runs the import as a background process and shows a notification when it completes; View Processes opens the Process Watcher with the command and its output.

Check the result with SELECT * FROM customers; in the Query Tool. In the sample, Dara Murphy's empty email arrives as NULL, because an unquoted empty value is the default null string in CSV format.

Import options explained

Where the CSV file must be (desktop vs server mode)

pgAdmin uses its own file dialog rather than the operating system's. In desktop mode it browses your local files. In server mode, which includes the Docker image and web installs, files are read from and stored on the pgAdmin server, so a CSV on your laptop is not visible until you upload it. Open Tools > Storage Manager, use Upload to put the file in your storage directory, then pick it in the Filename field.

Alternative: COPY and \copy in SQL

For repeatable imports, use SQL. COPY ... FROM 'file' reads a file on the database server, so the server must be able to see it and you need superuser rights or the pg_read_server_files role. The psql meta-command \copy reads the file from the client instead; you can run it in pgAdmin's PSQL Tool (Tools menu). HEADER MATCH makes PostgreSQL check that the header names match the table columns, in order.

-- File on the database server (superuser or pg_read_server_files):
COPY customers (full_name, email, signup_date)
FROM '/var/lib/postgresql/import/customers.csv'
WITH (FORMAT csv, HEADER true);

-- File on your machine, from psql or the pgAdmin PSQL Tool (one line):
\copy customers (full_name, email, signup_date) FROM 'customers.csv' WITH (FORMAT csv, HEADER true)

Common import errors and fixes

• A value does not fit the column type. A value does not convert to the column's type, often a date in another format or a header row imported as data. Turn Header on, fix the data, or on PostgreSQL 17 and later set On Error to ignore to skip bad rows.
• Too many or too few columns. The CSV has a different number of columns from the selection on the Columns tab, or the delimiter is wrong.
• Garbled accented characters. Set Encoding to the file's real encoding.
• File not found in server mode. Upload it with the Storage Manager first.

To get data out again, see the pgAdmin export guide in the related pages below, and practise the SQL in the SQL beginner course.

Frequently asked questions

Sources

• pgAdmin 4 documentation: Import/Export Data Dialog
• pgAdmin 4 documentation: Menu Bar
• pgAdmin 4 documentation: Storage Manager
• pgAdmin: FAQ (file dialogs)
• PostgreSQL 18 documentation: COPY

Checked 8 October 2026.

How we research tool guides: our editorial method. CodeWithSQL earns nothing from the vendors mentioned.