Mail merge


A mail merge takes values from rows of a spreadsheet or other data source and inserts them into a template document. This lets you create a single "master" document (the template) from which you can generate many similar documents, each customized with the data being merged. The result is not necessarily used for mail or form letters, but can be for any purpose at all, such as generating a batch of customer invoices.

Mail merge has been around as long as there have been spreadsheets and word processors and is part of many business workflows today. The convention is to organize the data as one record per row, with the columns representing fields in the data, as shown in the following table:

1 Name Address Zone
2 UrbanPq 123 1st St. West
3 Pawxana 456 2nd St. South
4 etc...

The sample app on this page shows how you can use the Google Docs and Sheets (and Drive) APIs to abstract away the details of how mail merges are performed, protecting users from implementation concerns. More information on this sample can be found at the sample's open source repo.

Sample application

This sample app copies your master template then merges variables from your designated data source into each of the copies. To try the sample app, first set up your template:

  • Create a new Google Docs file. Choose whatever template you wish to use. (Our sample template uses Letter/Spearmint.)
  • Note the document ID, which is the string that follows document/d/ (see DOCUMENT_ID) in the URL:
  • Set the DOCS_FILE_ID variable in the code to that document ID.
  • Replace the contact information in the Doc with template placeholder variables that the app will merge with desired data.

Here is our sample letter template with placeholders to be merged with real data from a source such as a Google Sheet or plain text. Here's what that template looks like:

Next, choose either plain text or Google Sheets as your data source by setting the SOURCE variable. It defaults to plain text, meaning the sample data in the TEXT_SOURCE_DATA variable. To source the data from a Google Sheet, update the SOURCE variable to 'sheets' and point it to our sample (or yours) by setting the SHEETS_FILE_ID variable. Here is our Sheet so you get an idea of the format:

Try the app with our sample data, then adapt to your data and use case. The command-line application works like this:

  • Setup
  • Fetch the data from the data source
  • Loop through each row of data
    • Create a copy of the template
    • Merge the copy with the data
    • Output link to newly-merged document

All of the new merged letters also show up in the user's Google Drive. An example of a merged letter looks something like this:

Source code


import time

import google.auth
from googleapiclient.discovery import build
from googleapiclient.errors import HttpError

# Fill-in IDs of your Docs template & any Sheets data source
DOCS_FILE_ID = "195j9eDD3ccgjQRttHhJPymLJUCOUjs-jmwTrekvdjFE"
SHEETS_FILE_ID = "11pPEzi1vCMNbdpqaQx4N43rKmxvZlgEHE9GqpYoEsWw"

# authorization constants

SCOPES = (  # iterable or space-delimited string

# application constants
SOURCES = ("text", "sheets")
SOURCE = "text"  # Choose one of the data SOURCES
COLUMNS = ["to_name", "to_title", "to_company", "to_address"]
        "Ms. Lara Brown",
        "Google NYC",
        "111 8th Ave\nNew York, NY  10011-5201",
        "Mr. Jeff Erson",
        "Google NYC",
        "76 9th Ave\nNew York, NY  10011-4962",

# fill-in your data to merge into document template variables
merge = {
    # sender data
    "my_name": "Ayme A. Coder",
    "my_address": "1600 Amphitheatre Pkwy\nMountain View, CA  94043-1351",
    "my_email": "",
    "my_phone": "+1-650-253-0000",
    # - - - - - - - - - - - - - - - - - - - - - - - - - -
    # recipient data (supplied by 'text' or 'sheets' data source)
    "to_name": None,
    "to_title": None,
    "to_company": None,
    "to_address": None,
    # - - - - - - - - - - - - - - - - - - - - - - - - - -
    "date": time.strftime("%Y %B %d"),
    # - - - - - - - - - - - - - - - - - - - - - - - - - -
    "body": (
        "Google, headquartered in Mountain View, unveiled the new "
        "Android phone at the Consumer Electronics Show. CEO Sundar "
        "Pichai said in his keynote that users love their new phones."

creds, _ = google.auth.default()
# pylint: disable=maybe-no-member

# service endpoints to Google APIs

DRIVE = build("drive", "v2", credentials=creds)
DOCS = build("docs", "v1", credentials=creds)
SHEETS = build("sheets", "v4", credentials=creds)

def get_data(source):
  """Gets mail merge data from chosen data source."""
    if source not in {"sheets", "text"}:
      raise ValueError(
          f"ERROR: unsupported source {source}; choose from {SOURCES}"
    return SAFE_DISPATCH[source]()
  except HttpError as error:
    print(f"An error occurred: {error}")
    return error

def _get_text_data():
  """(private) Returns plain text data; can alter to read from CSV file."""

def _get_sheets_data(service=SHEETS):
  """(private) Returns data from Google Sheets source. It gets all rows of
  'Sheet1' (the default Sheet in a new spreadsheet), but drops the first
  (header) row. Use any desired data range (in standard A1 notation).
  return (
      .get(spreadsheetId=SHEETS_FILE_ID, range="Sheet1")
  # skip header row

# data source dispatch table [better alternative vs. eval()]
SAFE_DISPATCH = {k: globals().get(f"_get_{k}_data") for k in SOURCES}

def _copy_template(tmpl_id, source, service):
  """(private) Copies letter template document using Drive API then
  returns file ID of (new) copy.
    body = {"name": f"Merged form letter ({source})"}
    return (
        .copy(body=body, fileId=tmpl_id, fields="id")
  except HttpError as error:
    print(f"An error occurred: {error}")
    return error

def merge_template(tmpl_id, source, service):
  """Copies template document and merges data into newly-minted copy then
  returns its file ID.
    # copy template and set context data struct for merging template values
    copy_id = _copy_template(tmpl_id, source, service)
    context = merge.iteritems() if hasattr({}, "iteritems") else merge.items()

    # "search & replace" API requests for mail merge substitutions
    reqs = [
            "replaceAllText": {
                "containsText": {
                    "text": "{{%s}}" % key.upper(),  # {{VARS}} are uppercase
                    "matchCase": True,
                "replaceText": value,
        for key, value in context

    # send requests to Docs API to do actual merge
        body={"requests": reqs}, documentId=copy_id, fields=""
    return copy_id
  except HttpError as error:
    print(f"An error occurred: {error}")
    return error

if __name__ == "__main__":
  # get row data, then loop through & process each form letter
  data = get_data(SOURCE)  # get data from data source
  for i, row in enumerate(data):
    merge.update(dict(zip(COLUMNS, row)))
        "Merged letter %d:"
        % (i + 1, merge_template(DOCS_FILE_ID, SOURCE, DRIVE))

For more information, see the README as well as the full application source code at this sample app's open source repo.