How Fusion Translates Data

Fusion translates data in two ways. 

The first type refers to formatting. In this scenario, a date field may need to be formatted consistently or a select field transformed from an ID value to a field value (e.g. country id to full country label). In both examples, data types are standardized so values that refer to same entity are put in a single, common format. To do this, Fusion finds the primary key fields (e.g. Company: {keys: ['COMPANY_NAME']} or Contact: {keys: ['EMAIL']}) then normalizes the field values to transform dates, primary key fields, and select/multiselect fields by using the schema information for each field. 

In the second scenario, a relationship field and the information from the related object both need to be pulled into a query (e.g. viewing the `company` field on a Contact record as the name of the company instead of the company ID). To do this, Fusion builds links tables. With the links tables, you can combine two tables (e.g. Contact and Company) in SQL with a single query. So if you wanted to join fused_contact and fused_company, you would do so via the contact_company_links table. Similarly, you could join the fused_opportunity and fused_contact table via the links_contact_opportunity table. For example:

SELECT fused_contact.hubspot_score AS `Score`,
       fused_opportunity.deal_stage AS `Stage`,  
       fused_opportunity.close_date AS `Date`  
FROM fused_opportunity  
JOIN links_contact_opportunity ON fused_contact.contact_id = links_contact_opportunity.contact_id

In short, Fusion also translates data by letting you join two or more tables (e.g. Contact and Company) together by building a SQL query with the relevant links tables (e.g. contact_company_links).

Still need help? Contact Us Contact Us