If I have a single data source, how does using the Fused tables help (vs. using the Raw tables)?
While there is a lot of value to connecting more than one data source, here are a few ways Fusion is valuable with a single data source.
1. Data deduplication. Fusion eliminates redundant values in your dataset, removing extra copies and storing only one unique value.
2. Type Translation. All field values (date fields, emails, etc.) and their concomitant data types are standardized. Values that refer to same entity are put in a single, common format.
3. Links and Contributors tables. Both allow you to join tables together more easily. While the links tables form relationships between fused objects (Contact, Company, Opportunity, List, Event, Owner, Ticket, Activity, Product, Order), the contributors tables link raw records to fused records (e.g. going from hubspot0_contact to fused_contact in a single query).
4. SQL access to raw data. Query raw data in any BI tool that accepts MySQL.
How often is my Fusion Warehouse refreshing?
Approximately every 30 minutes we will pull in updates from your systems.
Important Note: it's possible during big updates in your systems that the timing between refreshes can be much longer than 30 minutes.
Does this mean my reports will be updated every 30 minutes too?
This varies depending on what solution you have chosen for visualizing your Fusion database. For some of the lighter-weight / free options a common timeframe is to update the reports every hour with refreshed data whereas the more enterprise-type solutions will refresh data in real time.
What database types does Fusion support?
We support MySQL. Support for additional databases is coming soon.
Can you tell me more about the warehouse's technical specifications?
Fusion’s cloud data warehouse is a hosted MySQL solution. As a relational SQL data warehouse built for the cloud, Fusion is hosted on AWS and runs on Amazon RDS.
Thanks to this MySQL architecture, Fusion provides total relational database support for structured columnar stores and comprehensive support for the SQL language. And because Fusion is an acceptable data source for any service that accepts MySQL, it also works great for developers looking to build modern data applications.
How does Fusion build my Fused Database?
Fusion starts by deduping records based on a specific field (see below for these default dedupe fields). Once we find records, we then pull in the fields based on the default field schemas and any custom fields you’ve added to your Fusion mappings. Once we have the field data, if you are mapping multiple fields together from the different systems, we choose whatever system was last modified and use that systems value for this mapped field.
During the field building process, we are normalizing the data types such as the following:
- Select-type fields (i.e. drop down select, option set)
- Boolean fields (i.e. single checkbox, yes/no)
- Number fields
- Date and Date/Time fields
- Text fields (i.e. single line text, text area)
During the normalization of field types we are also structuring the field data in a normalized manner to ensure accuracy when building your reports.
We then look at the relationships that each record in a Fused record set has within their respective systems and list these related records in the Fused record set (i.e. Contact to Company relationship, Contact to Opportunity, Contact to Activity, etc.).
How does Fusion Deduplicate Data to build my Fused Database?
By default we use the following fields / relationships to match & dedupe records:
|Object||Identifying Fields||Identifying Relationships||Matching Logic|
|Contact||Direct match on lowercased email address|
|Company||Company_Name||Fuzzy match on company name; Long term this will be come a more complex soft match|
|Opportunity||Name, Close_Date||Contact and/or Company||Fuzzy match on name, direct match on expected close date, and graph match on related contact/company|
|List||Name||Fuzzy match on name|
|Event||Name, Start_Time||Fuzzy match on name; direct match on start time|
|Owner / Employee||Direct match on lowercased email address.|
|Support Ticket||Name, Number||Contact and/or Company||Fuzzy match on name, direct match on number, and graph match on related contact/company|
|Activity||Title, Occured_At||Contact and/or Company||Direct match on title and "occured at", and graph match on related contact/company|
|Product||SKU||Direct match on SKU|
|Order||Occured_At||Contact and/or Company||Direct match on "occurred at" date/time, and graph match on related contact/company|
At a later date we'll be releasing the ability to customize these dedupe keys, so stay tuned.
How does Fusion Calculate my record to determine my plan?
We use the raw data tables for each system and add up the total number of records within each raw data table.
How does Fusion know what system to use for updating fields for my Fused records?
We use a "Last Modified" process where we'll use whatever system was last modified to update the respective Fused record fields.
What browsers are supported by Fusion?
We support the latest versions of Chrome, Firefox and Safari.
We do not support Internet Explorer 11 at this time.
Existing Sync Customers FAQs
I’m an existing customer using the Sync product, can I sign up using my current installed Connectors?
I’m an existing customer using both Sync and Fusion, can I use my existing mappings to build my Fused database?
Yes, however, if you make changes to your mapped fields in your Sync portal, this will require a refresh of your Fused Database, which can take any where from an hour to a few hours depending on your systems and database size.
I’m an existing customer using both Sync and Fusion and I’m going to be on the paid version of Fusion, will I get billed at the same time or billed separately?
You will be billed separately.