Skip to content


Switch branches/tags
This branch is 14 commits ahead, 38 commits behind 18F:master.

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time

Code Climate CircleCI Dependency Status

Analytics Reporter

A lightweight system for publishing analytics data from Google Analytics profiles. Uses the Google Analytics Core Reporting API v3 and the Google Analytics Real Time API v3.

This is used in combination with 18F/ to power the government analytics hub,

Available reports are named and described in reports.json. For now, they're hardcoded into the repository.

<<<<<<< HEAD


Windows servers

Install node.js for Windows

Install JavaScript dependencies (from the directory containing the local copy of this github repository):

npm install

Linux servers

See analytics-reporter for Linux setup instructions.

Google Analytics


  • Create an API service account in the Google developer dashboard.

  • Visit the "APIs" section of the Google Developer Dashboard for your project, and enable it for the "Analytics API".

  • Go to the "Credentials" section and generate "service account" credentials using a new service account.

  • Download the JSON private key file it gives you.

  • Grab the generated client email address (ends with from the contents of the .json file.

<<<<<<< HEAD

  • Add a system environment variable to specify the path to the private key file you downloaded in the previous step:
  • Take the generated client email address (ends with and grant it Read, Analyze & Collaborate permissions on the Google Analytics profile(s) whose data you wish to publish.

  • Add system environment variables for your app's generated email address, and for the profile you authorized it to:


To find your Google Analytics view ID:

  1. Sign in to your Analytics account.
  2. Select the Admin tab.
  3. Select an account from the dropdown in the ACCOUNT column.
  4. Select a property from the dropdown in the PROPERTY column.
  5. Select a view from the dropdown in the VIEW column.
  6. Click "View Settings"
  7. Copy the view ID. You'll need to enter it with ga: as a prefix.
  • You can specify your private key through environment variables either as a file path, or the contents of the key (helpful for Heroku and Heroku-like systems).

To specify a file path (useful in development or Linux server environments):

export ANALYTICS_KEY_PATH="/path/to/secret_key.json"

Alternatively, to specify the key directly (useful in a PaaS environment), paste in the contents of the JSON file's private_key field directly and exactly, in quotes, and rendering actual line breaks (not \n's) (below example has been sanitized):

[contents of key]

If you have multiple accounts for a profile, you can set the ANALYTICS_CREDENTIALS variable with a JSON encoded array of those credentials and they'll be used to authorize API requests in a round-robin style.

    "key": "-----BEGIN PRIVATE KEY-----\n[contents of key]\n-----END PRIVATE KEY-----",
    "email": ""
    "key": "-----BEGIN PRIVATE KEY-----\n[contents of key]\n-----END PRIVATE KEY-----",
    "email": ""
  • Make sure your computer or server is syncing its time with the world over NTP. Your computer's time will need to match those on Google's servers for the authentication to work.

  • Test your configuration by printing a report to STDOUT:

node bin/analytics --only realtime

If you see a nicely formatted JSON file, you are all set.

  • (Optional) Authorize yourself for S3 publishing.

If you plan to use this project's lightweight S3 publishing system, you'll need to add 6 more environment variables:

export AWS_REGION=us-east-1
export AWS_ACCESS_KEY_ID=[your-key]
export AWS_SECRET_ACCESS_KEY=[your-secret-key]

export AWS_BUCKET=[your-bucket]
export AWS_BUCKET_PATH=[your-path]

There are cases where you want to use a custom object storage server compatible with Amazon S3 APIs, like minio, in that specific case you should set an extra env variable:

export AWS_S3_ENDPOINT=http://your-storage-server:port

Other configuration

If you use a single domain for all of your analytics data, then your profile is likely set to return relative paths (e.g. /faq) and not absolute paths when accessing real-time reports.

You can set a default domain, to be returned as data in all real-time data point:


This will produce points similar to the following:

  "page": "/post/why-google-is-hurrying-the-web-to-kill-sha-1",
  "page_title": "Why Google is Hurrying the Web to Kill SHA-1",
  "active_visitors": "1",
  "domain": ""


Reports are created and published using the analytics command.

node analytics

This will run every report, in sequence, and print out the resulting JSON to STDOUT. There will be two newlines between each report.

A report might look something like this:

  "name": "devices",
  "query": {
    "dimensions": [
    "metrics": [
    "start-date": "90daysAgo",
    "end-date": "yesterday",
    "sort": "ga:date"
  "meta": {
    "name": "Devices",
    "description": "Weekly desktop/mobile/tablet visits by day for all sites."
  "data": [
      "date": "2014-10-14",
      "device": "desktop",
      "visits": "11495462"
      "date": "2014-10-14",
      "device": "mobile",
      "visits": "2499586"
      "date": "2014-10-14",
      "device": "tablet",
      "visits": "976396"
    // ...
  "totals": {
    "devices": {
      "mobile": 213920363,
      "desktop": 755511646,
      "tablet": 81874189
    "start_date": "2014-10-14",
    "end_date": "2015-01-11"


  • --output - Output to a directory.
node analytics --output /path/to/data

Note that when using the docker image you have to use the absolute path, for example "/home/youruser/path/to/data"

  • --publish - Publish to an S3 bucket. Requires AWS environment variables set as described above.
node analytics --publish
  • --write-to-database - write data to a database. Requires a postgres configuration to be set in environment variables as described below.

  • --only - only run one or more specific reports. Multiple reports are comma separated.

node analytics --only devices
node analytics --only devices,today
  • --slim -Where supported, use totals only (omit the data array). Only applies to JSON, and reports where "slim": true.
node analytics --only devices --slim
  • --csv - Gives you CSV instead of JSON.
node analytics --csv
  • --frequency - Limit to reports with this 'frequency' value.
node analytics --frequency=realtime
  • --debug - Print debug details on STDOUT.
node analytics --publish --debug

Saving data to postgres

The analytics reporter can write data is pulls from Google Analytics to a Postgres database. The postgres configuration can be set using environment variables:

export POSTGRES_HOST = ""
export POSTGRES_USER = "postgres"
export POSTGRES_PASSWORD = "123abc"
export POSTGRES_DATABASE = "analytics"

The database expects a particular schema which will be described in the API server that consumes this data.

To write reports to a database, use the --write-to-database option when starting the reporter.

Deploying to GovCloud

The analytics reporter runs on ☁️.gov. Please refer to the manifest.yml file at the root of the repository for application information.

Ensure you're targeting the proper org and space.

cf target

Deploy the application with the following command.

cf push -f manifest.yml

Set the environmental variables based on local .env file.

cf set-env analytics-reporter AWS_ACCESS_KEY_ID 123abc
cf set-env analytics-reporter AWS_SECRET_ACCESS_KEY 456def
# ...

Restage the application to use the environment variables.

cf restage analytics-reporter

Developing with Docker

This repo contains a Docker Compose configuration. The reporter is configured to run in the container as if it were running in GovCloud. This is helpful for seeing how the reporter will behave when deployed without pushing it to

To start the reporter, first run the docker-update script to install the necessary dependencies:


Note that this script will need to be run again when new dependencies are added to update the Docker volumes where the dependencies are stored.

After the dependencies are installed, the reporter can be started using Docker Compose:

docker-compose up

Public domain

This project is in the worldwide public domain. As stated in CONTRIBUTING:

This project is in the public domain within the United States, and copyright and related rights in the work worldwide are waived through the CC0 1.0 Universal public domain dedication.

All contributions to this project will be released under the CC0 dedication. By submitting a pull request, you are agreeing to comply with this waiver of copyright interest.


Lightweight analytics reporting and publishing tool for Google Analytics data. Powers,,,, and more.







No packages published


  • JavaScript 80.5%
  • Shell 19.4%
  • Makefile 0.1%