Data requirements

(REQUIRED) Usage events, entity profiles, and conversion events

1. Users or group profiles (whatever entity is associated with conversion and upsell events)

   1. A table of all your users in a data warehouse, a table of all your accounts, Segment identify calls, etc
   2. Association between entities i.e. this user belongs to this workspace and this workspace belongs to this account, this contract belongs to this account
      • e.g. Segment Group calls
      • e.g. a lookup table of user IDs and group IDs
   3. (Optional) It’s ideal to have historical profile information about users, accounts, etc so that we know how attributes about these objects change over time.

2. Product usage activity events: what the users or groups do in your product, usually tracked by tools like Segment, Heap, or Amplitude

   1. Minimum fields needed: Primary key, Timestamp, Event Name, ID to associate with user/group

3. Revenue events (conversion/upsell events): when and which accounts converted or upsold, usually found in CRM as opportunities/deal or billing data as transactions. Specifically:

   1. Conversions from free to paid
   2. Upsells onto higher tier
   3. Contraction/churn

   Requirements

   • We need at least 50 conversion events to run data science on your data.
   • A field to link revenue data sets to object data

(Optional) Nice-to-have data

Accessing your data: how we connect to your data

We connect to your data in 4 ways:

1. Data warehouses such as Redshift, Snowflake, and BigQuery (requires sharing credentials to access data warehouse)
   1. HeadsUp runs queries in customer’s data warehouse and requires credentials to the warehouse.
   2. Customer must provide read user credentials to the data warehouse.
2. Streaming event sources like Segment (no engineering lift; requires setting up HeadsUp as a destination in the sources admin panel)
   1. HeadsUp receives events directly from Segment or another Webhook and stores it in our own data warehouse.
   2. Customer must OAuth into Segment and configure HeadsUp as a destination (takes 1-2 minutes).
3. Application sources like Salesforce, Stripe, etc (no engineering lift; require user credentials to these applications)
   1. HeadsUp will pull data directly from these sources using ETL and store it in our own data warehouse.
   2. Customer must provide API token or walk through an OAuth flow.
   3. For BigQuery: BigQuery service account with BigQuery User and BigQuery Data Viewer roles
4. Flat files
   1. HeadsUp can also ingest flat files such as CSVs

Details on data requirements

This is an example of the tables we need:

• 🔗 https://docs.google.com/spreadsheets/d/1QnOpYH4CX5zDLqlMH8xcq0H96F55QPqUqzY6VlTDru0/edit?usp=sharing