Stratifyd’s Data Fusion capabilities let users join or “fuse” different kinds of data from different kinds of data sources inside the platform. Doing so consolidates information into an “apples-to-apples” analytical context so that users can apply one single AI-driven model to the whole group. Analyzing data in this way allows users to gain a more complete view of customer experiences across channels and sources (surveys, calls, chats, support tickets, etc.)


While the ability to conjoin different types of data from different sources within the platform provides for easier analysis, there are several practices that we suggest in order to mitigate the effects of the feature’s limitations.

1. Data Fusion allows for the joining of a maximum of two data streams.


This feature is technically capable of joining more than three data streams, but in order to do so, multiple ETL streams must be created (e.g. joining stream 1 and 2 to create ETL 1, and then joining ETL 1 & stream 3 to create ETL 2).

NOTE: This limitation is expected to be resolved in future version of the Data Fusion feature.

2. Data Fusion is only available for static data streams.


In order to fuse data from different streams, those streams must not have continually changing/updating data, nor can they be continually acquiring new data.

NOTE: This limitation is expected to be resolved in future version of the Data Fusion feature.

3. Data Fusion requires the use of the SQL editor.


In order to fuse data from different streams, the user must execute functions using SQL. For more information about SQL, visit the APPENDIX (below).

NOTE: Future versions of the Data Fusion feature are expected to include a user interface that offers visual point-and-click functionality.

4. Users must select two streams, though Data Fusion can be used on just one stream.


To use the Data Fusion feature, the users must select two data streams to “fuse”. However, they do have the option to modify/alter/group/conjoin the contents of a single data stream, like renaming columns, adding calculations, grouping data, etc., if need be.

5. Cannot store temporary tables when using Data Fusion.


SQL typically allows for storing results in a temporary table for later reference, but that action is not available in this version of the Data Fusion feature without the use of subquery options.


SQL is a programming language designed to help manage data, especially structured data, within databases. When using the Data Fusion feature, users will need to use one of several important functions in order to capture the information they want. Overall, there are three basic functions in an SQL query. Below is a brief description of each function.

SELECT – Input field(s) that you wish to have returned in the query.

FROM – Input table name that you wish to pull data from.

WHERE – Insert any filters that are needed.


Let’s imagine I have a data set made up of information regarding the weather. Then, imagine that I want to pull specific information from the data set, like the temperature, wind, and precipitation for each day over 60 degrees. The appropriate SQL functions I might use are listed below.



, WeatherDataTable.Wind

, WeatherDataTable.Date

, WeatherDataTable.Precipitation




WeatherDataTable.Temperature_Farenheit > 60

Other SQL Considerations…

Additional information regarding SQL commands and/or practices which may be of value when using the Data Fusion feature is listed below.

  • For most accurate results, reference the data table and field name. For instance, in the weather example (above), the separator is “.” (e.g. WeatherDataTable.Wind)
  • To keep your tables and fields sorted, you can rename them using the function “as” For instance, to rename a table: WeatherDataTable as Weather
  • To select all fields, use an asterisk (i.e. “*”)
  • To select specific fields, use a comma (“,”) as a separator. For instance, the weather example (above), the comma separator precedes each individual field (i.e. WeatherDataTable.Wind , WeatherDataTable.Date , WeatherDataTable.Precipitation). NOTE: A comma separator is not needed for the first specified field, only subsequent ones.

For more information, visit:

Did this answer your question?