Overview

Reforge Insights can connect to and pull data from your data warehouse of choice. Currently, we support Snowflake and BigQuery, but are always looking to expand our integrations.

Supported Warehouses

Snowflake

Secure Data Sharing (Preferred)

To connect with Snowflake, you can request our account details and use Secure Data Sharing. Secure data sharing allows you to give us the targeted access to just the data you want to share.

Service User (Alternative)

If setting up secure data sharing is not an option, you can instead create a service user with a password and share the credentials with us. Create a role for the service user and grant it read access on the tables you wish to connect. Then share the following with us:

account url
username
password
role
warehouse
database

We connect to Snowflake using the following IP addresses, which you may need to whitelist in your Snowflake account.

34.212.54.206
35.162.156.254
44.241.212.223
44.240.0.247

BigQuery

Amazon Redshift

To connect with Amazon Redshift using Datashare:

Create a datashare in your Redshift cluster:

CREATE DATASHARE ingest_share;

Add objects to your datashare. You can add entire schemas or specific tables and views:

-- Whole schema
ALTER DATASHARE ingest_share ADD SCHEMA public;

-- Or pick specific items
ALTER DATASHARE ingest_share ADD TABLE public.conversations;
ALTER DATASHARE ingest_share ADD VIEW public.latest_tickets_v;

Make the datashare visible cross-account:

ALTER DATASHARE ingest_share SET PUBLICACCESSIBLE TRUE;

Note: This doesn’t expose your data publicly. It only allows consumption by authorized accounts.

Authorize Reforge’s AWS account to access your datashare:

aws redshift authorize-data-share \
  --data-share-arn arn:aws:redshift:us-east-1:111122223333:datashare:ingest_share \
  --consumer-accounts 44445556677

Replace the ARN with your datashare ARN and the consumer-accounts value with the AWS account ID provided to you by your Reforge contact.

You can also do this via the Redshift console under Datashares → ingest_share → Authorize.

Provide the following information to your primary contact at Reforge:

Field	Example
Datashare ARN	arn:aws:redshift:us-east-1:111122223333:datashare:ingest_share
Redshift namespace ARN (only if you have multiple clusters/workgroups)	arn:aws:redshift:us-east-1:111122223333:namespace:prod-ns-1
List of schemas / tables / views included	public.conversations, public.latest_tickets_v

Data Models

We support several schemas and relationships between tables. With each of the schemas defined below, you can also provide a list of additional column names that you would like to be included on the respective fields as metadata. A few examples of common additional fields are for feedback or conversation messages, a “version” number that the feedback refers to, or for Contacts, a “plan” or “subscription tier”. When working with the team to connect to your data warehouse, simply share with us fields beyond those defined below that you would like to have included!

Standalone Feedback

Standalone feedback are singular items of feedback, such as a review or a single response to a question.

Field Name	Required	Type	Description
id	Yes	STRING	Unique identifier for the feedback record
description	Yes	STRING	The main feedback content
created_at	Yes	STRING	When the feedback was created
updated_at	Yes	DATE	When the feedback was last updated
source	No	STRING	The source of the feedback, e.g. “HUBSPOT”
title	No	STRING	Optional title for the feedback
contact_id	No	STRING	Optional identifier linking to a contact

Conversation

A conversation is a collection of conversational feedback messages. For example, a chat log between a support agent and a user.

Field Name	Required	Type	Description
id	Yes	STRING	Unique identifier for the conversation
title	No	STRING	Optional title for the conversation
source	No	STRING	The source of the feedback, e.g. “HUBSPOT”
created_at	Yes	STRING	When the conversation was created
updated_at	No	DATE	When the conversation was last updated

Conversation Message

These are the individual messages within a conversation, and should be linked to a conversation by the conversation_id field.

Field Name	Required	Type	Description
id	Yes	STRING	Unique identifier for the message
created_at	Yes	DATE	When the message was sent
updated_at	No	DATE	When the message was last updated
conversation_id	Yes	STRING	ID of the conversation this message belongs to
content	Yes	STRING	The message content
incoming	Yes	BOOLEAN	Whether the message was sent by an external user
contact_id	No	STRING	Optional identifier linking to a contact

Contacts

Contacts are the users of your product.

Field Name	Required	Type	Description
id	Yes	STRING	Unique identifier for the contact
name	Yes	STRING	The contact’s full name
email	Yes	STRING	The contact’s email address
account_id	Yes	STRING	The account the contact belongs to
created_at	Yes	DATE	When the contact was created
updated_at	Yes	DATE	When the contact was last updated

Accounts

Accounts are the companies that your contacts belong to.

Field Name	Required	Type	Description
id	Yes	STRING	Unique identifier for the account
name	Yes	STRING	The account’s name
domain	Yes	STRING	The account’s domain
contract_value	No	NUMBER	The account’s contract value
created_at	Yes	DATE	When the account was created
updated_at	Yes	DATE	When the account was last updated

Custom Tables

Sometimes, you may not be able to map your data to the existing schemas. In this case, you can share your own, existing tables. This will require custom engineering work on our end, so understand that time will be required to support this.

If you are providing custom tables, then what we will need to do is to effectively write custom queries and map the data to schemas above on our end. Given that, helping us understand how to do that mapping will expedite the process and lead to better, faster results for getting your data into Reforge Insights.

When sharing custom table information, please provide the following:

1. Table Information The table names, column names, column types, and a description of the data in each column.

For example:

Customer Table: client_organizations

Column Name	Column Type	Description
org_id	STRING	The unique identifier for the organization
org_name	STRING	The name of the organization (maps to account name)
org_website	STRING	The organization’s website domain (maps to account domain)
annual_revenue	NUMBER	The organization’s annual contract value
creation_date	DATE	When the organization was created (maps to account created_at)
last_modified	DATE	When the organization was last updated (maps to account updated_at)

Customer Table: client_users

Column Name	Column Type	Description
user_id	STRING	The unique identifier for the user (maps to contact id)
full_name	STRING	The user’s full name (maps to contact name)
user_email	STRING	The user’s email address (maps to contact email)
org_id	STRING	Foreign key to client_organizations (maps to account_id)
signup_date	DATE	When the user was created (maps to contact created_at)
profile_updated	DATE	When the user profile was last updated (maps to contact updated_at)

Customer Table: support_tickets

Column Name	Column Type	Description
ticket_id	STRING	Unique identifier for the support ticket (maps to conversation id)
opened_at	DATE	When the ticket was created (maps to conversation created_at)
closed_at	DATE	When the ticket was closed (maps to conversation updated_at)
user_id	STRING	User who opened the ticket (maps to contact_id)
org_id	STRING	Organization the ticket belongs to
status	STRING	Current status of the ticket
priority	STRING	Priority level of the ticket

Customer Table: ticket_messages

Column Name	Column Type	Description
message_id	STRING	Unique identifier for the message (maps to message id)
ticket_id	STRING	Foreign key to support_tickets (maps to conversation_id)
message_text	STRING	The content of the message (maps to message content)
sent_time	DATE	When the message was sent (maps to message created_at)
edited_time	DATE	When the message was edited (maps to message updated_at)
from_customer	BOOLEAN	Whether the message was from the customer (maps to incoming)
user_id	STRING	User who sent the message (maps to contact_id)

2. Relationship Information If there are any relationships between the tables, please point out how the tables should be joined, as in which column on table A should be joined with which column on table B.

3. Data Mapping An explanation of how you would expect the data to map to the schemas defined above.

If you are interested in sharing custom tables, please reach out to your primary contact at Reforge.

Social media mentions & reviews

Customer support & CRM

Survey & feedback collection

Task & project management

Survey

Other Integrations

Data Warehouse

Overview

Supported Warehouses

Secure Data Sharing (Preferred)

Service User (Alternative)

Data Models

Standalone Feedback

Conversation

Conversation Message

Contacts

Accounts

Custom Tables

Social media mentions & reviews

Customer support & CRM

Survey & feedback collection

Task & project management

Survey

Other Integrations

​Overview

​Supported Warehouses

​Data Models

​Standalone Feedback

​Conversation

​Conversation Message

​Contacts

​Accounts

​Custom Tables

Overview

Supported Warehouses

Data Models

Standalone Feedback

Conversation

Conversation Message

Contacts

Accounts

Custom Tables