How is exported analytics data structured?

First, a few notes about the raw analytics export process and data:

  • Data is exported every 24 hours in CSV format. In the rare event of maintenance or downtime, the export will be delayed and eventually catch up to regular intervals.
  • Exported data will be synced from the day you first signed up, not from the day you enabled the export.
  • All times are in UTC, regardless of your timezone settings on the Account page (you get data in the exact format we store it).
  • Data is exported every 24 hours. If there's no analytics events for that day, an empty file is still created.
  • If you change your bucket settings mid-way, the data will be written to the new bucket without re-populating previous days.
  • We keep track of exported data locally, so if you delete exported files in your bucket, they will not be re-populated automatically.

Here's a sample listing of exported analytics data:

https://s3.amazonaws.com/my-website-s3-bucket/2015-10-29.csv
https://s3.amazonaws.com/my-website-s3-bucket/2015-10-30.csv
https://s3.amazonaws.com/my-website-s3-bucket/2015-10-31.csv
https://s3.amazonaws.com/my-website-s3-bucket/manifest.csv

The files ending with dates contain raw analytics events for that day (in UTC timezone). Please note that each row does not represent visitors, but rather visitor events:

visitor_id,event_type,lat,lng,address,visitor_created,event_created,event_data
84335a75202ed14cae6cd2d275df6b87,MAP_LOADED,38,-97,,1439160795,1439153192,"{""address"":""San Francisco, CA"",""url"":""http://first-example.com/page1""}"
84335a75202ed14cae6cd2d275df6b87,LOCATION_OPENED,38,-97,,1439160795,1439153193,"{""address"":""1390 Market Street, San Francisco, CA 94102"",""fullAddress"":""Coffee Shop, 1390 Market Street, San Francisco, CA 94102"",""locationId"":11404,""isLocationOpen"":false}"
84335a75202ed14cae6cd2d275df6b87,LOCATION_OPENED,38,-97,,1439160795,1439153195,"{""address"":""3995 24th Street, San Francisco, CA 94114"",""fullAddress"":""Coffee Shop, 3995 24th Street, San Francisco, CA 94114"",""locationId"":11383,""isLocationOpen"":false}"
0c83bddd530ceeec9d5b9b555ae07067,MAP_LOADED,40.721355,-74.000741,,1439160795,1439160795,"{""address"":""New York, NY"",""url"":""http://second-example.com/page2""}"
0c83bddd530ceeec9d5b9b555ae07067,LOCATION_OPENED,38,-97,,1439160795,1439160797,"{""address"":""1390 Market Street, San Francisco, CA 94102"",""fullAddress"":""Coffee Shop, 471 Broadway, New York, NY 10013"",""locationId"":11404,""isLocationOpen"":false}"

In the sample CSV file above there are two visitors with a total of 5 events: the first has loaded the widget and opened two locations, and the other has loaded the widget and opened one location. Each event contains several properties:

  1. visitor_id — this is a unique id assigned to each locator visitor.
  2. event_type — the type of event this entry represents:

    • MAP_LOADED — initial map load event
    • LOCATION_OPENED — when user clicks on a location, or one is automatically opened after search
    • STREET_VIEW_OPENED — when street view modal is opened
    • HOURS_OPENED — user clicked on the hours sidebar
    • DIRECTIONS_LOOKUP — user looked up directions to a location
    • ADDRESS_CHANGE — user did a search (either by typing, using browser geolocation or voice recognition)
    • PHONE_NUMBER — user clicked on a phone number (does not necessarily mean they dialed the number)
    • EMAIL_CLICKED — user clicked on the email
    • WEBSITE_CLICKED — user clicked on the website button
  3. lat and lng — geographical coordinates for the user. Sometimes we're unable to accurately geolocate the user, so you might see coordinates like 38,-97 which are country-level coordinates.
  4. address — the geolocated city of the user when they first opened the locator (if we were able to reverse-geocode it).
  5. visitor_created — a unix timestamp of when the visitor profile was first created.
  6. event_created — a unix timestamp of when the event was first created.
  7. event_data — any additional data for the event in JSON format.

To help you parse numerous files that appear in the bucket, we also publish a manifest.csv file:

date,events,minimum_timestamp,maximum_timestamp,url
2015-10-29,12345,1446076800,1446163199,https://s3.amazonaws.com/my-website-s3-bucket/2015-10-29.csv
2015-10-30,123,1446163200,1446249599,https://s3.amazonaws.com/my-website-s3-bucket/2015-10-30.csv
2015-10-31,4567,1446249600,1446335999,https://s3.amazonaws.com/my-website-s3-bucket/2015-10-31.csv
  1. date — the date of events in UTC timezone.
  2. events — the number of events within the file.
  3. minimum_timestamp and maximum_timestamp — the minimum and maximum timestamps of the data found within the file in UNIX format.
  4. url — complete url for the CSV file.
Was this article helpful? /
How can we improve this page?
This feedback is used only to improve this support page. Please email us directly at support@zenlocator.com if you'd like assistance.