Setting Up Your Data Import

Choose an import type:

HockeyStack has four primary types of data for the purposes of a data import. Aligning your source data to our internal data models helps us properly connect what you're importing to the data that already exists in HockeyStack.

Type	Description
Website Actions	Each row represents a website action taken by a user or company. This is data tracked separately from our website tracker script. It might be older datapoints before you integrated HockeyStack or secondary websites.
Custom Actions	Each row represents any kind of action taken by a user or company. That means anything that falls outside of website data. Most common examples are trial data for PLG customers or product analytics data that HockeyStack does not currently integrate with.
Properties	Objects and their properties, usually relating to accounts, users and deals. Although we pull all kinds of objects from CRMs and other GTM tools, you might have additional datapoints you want to connect with them. Most common examples include subscription data that cannot be tracked through the CRM or product line items that are updated through the data warehouse directly. This data type also supports importing additional properties for objects we already track - like importing additional properties for accounts in your CRM. Each row has an object type (e.g. subscription), unique object id (e.g. subscription id), and the object's corresponding properties (e.g, plan name, subscription start date etc.)
Metadata	Aggregated metrics that can't be tied to a single person or a company but still include a timestamp. Most prominent example currently supported is Campaign / Ad data, but this can also include other marketing activity that can be described as such. There are three metrics that can be tracked: spend, clicks, or impressions. Each row has a timestamp and the combination of campaign, ad group, ad, and UTM information.

How to best structure your data for an import

Each data type described above has a schema and required fields to ensure that the data you import can be properly connected to the data in HockeyStack. After choosing an import type, please ensure that the schema within your data warehouse matches the schema outlined in one of the import types listed below:

Import Website ActionsImport Custom ActionsImport PropertiesImport Metadata

Overall Data Requirements:

One Time vs Recurring Syncs

• One Time syncs import data from your source into HockeyStack only once. This primarily applies to historical data that will not change over time.

• Recurring syncs will do a daily import of the updated information from your data warehouse to HockeyStack.

  • For S3 imports, the process is based on the file's upload time. This means that for each import type, new data should be included in a new file.

  • For all other providers, new data is pulled based on a timestamp from a specified column (such as last_added_date). This timestamp should reflect when the data was inserted into your source since it will be used to distinguish what data we should be pulling from the source.

Identifier Column

Website Actions, Custom Actions and Properties import types require an identifier column associated with each row of data. This column signifies who or what entity this row is tied to. For user-based data, this is generally an email address and for company-based data this is a company domain. Even for custom objects like subscriptions you would still have an email or an account attached to this subscription.

• To ensure that a record is imported properly, this identifier needs to be present, or else there is no way to connect this record to other data within the platform.

• The only exception to an email address or company domain is another unique identifier for a record provided by a CRM (e.g. Salesforce ID)

• Records that lack this requirement can not be imported and will be missing from the import.