Data Feed Guide: SIS Integration

Overview

This article outlines rules, tips, and recommendations for building data feeds using any method, as well as for specific methods and implementations. There are also some best practices for how to set up specific feeds and fields found in the SS&E - SIS Integration section of the Help Center.

All Methods

Requirements for feeds and Accelerate configuration especially for new implementations.

1. Ids must be unique

   Note: Many times, this Id is a combination of values imported on other columns that together create a unique identifier. For example, on the Person Financial Aid Indicator data feed, you may use a combination of the financial aid indicator id, person id, and term id to create a unique Id, which can be similarly done on other data feeds.

2. Any field that references another data feed (field name usually ends in "Id") must match the id of a record in the other feed. See hidden column "referenced in" in your Data Sheet to see a list of feeds that reference each feed.
3. All person-related data feeds that reference a personId should be limited to only include records that are also in the person feed (ex: personApplication, personAddress, transcript). This applies to all associated feeds, but is most common in person-related data.
4. All date formats must be YYYY-MM-DD, YYYY-MM-DD HH:mm:ss, or YYYY-MM-DDT00:00:00-00:00 with timezone offset. Example: "2025-01-01T00:00:00-05:00" for EST.
5. Cannot use <> in Accelerate configuration because of the XML format. Must substitute "&lt;" and "&gt;"
6. Comments in Accelerate configuration formatted: <!-- comment here -->
7. Can use static definitions if a list will be referenced throughout Accelerate in multiple queries.
8. Accelerate may be used to perform logic and manipulate/transform data after extraction (CSV) or in the extraction process (Direct DB)
9. Datatypes must be accurate
10. Less is more: data is much easier to add later than to delete once it has imported. If you are not sure about something, filter it out. Some types of records are easier than others to remove, but as a rule, if it's associated with any other records it's very difficult.
11. Start with a draft of each query and don't worry about getting everything perfect.
12. Include historical records for anything that will be on a student's transcript, courses, sections, degrees, programs, etc.

Cloud Connect

Using CSVs with Cloud Connect, hosted by Watermark.

1. Ids must be unique
   Note: Many times, this Id is a combination of values imported on other columns that together create a unique identifier. For example, on the Person Financial Aid Indicator data feed, you may use a combination of the financial aid indicator id, person id, and term id to create a unique Id, which can be similarly done on other data feeds.
2. CSV files must have headers
3. CSV files should use UTF-8 (or raw ASCII) encoding without BOM
4. Empty fields in CSV files may be set to "" or null
5. CSV files should be properly formatted to take into consideration the possibility of commas or double quotes in a data element.
6. Additional columns may be added to the CSV file and used for logic or calculations for other fields. Discuss with your Technical Consultant for custom fields.
7. Additional or separate files may be used and combined into one feed using Accelerate queries

Legacy Methods

These methods will remain available indefinitely, but new implementations will use Cloud Connect above.

Connect Method 1: CSV

Using CSVs with Connect2 hosted on the institution's Connect Server.

1. Ids must be unique

   Note: Many times, this Id is a combination of values imported on other columns that together create a unique identifier. For example, on the Person Financial Aid Indicator data feed, you may use a combination of the financial aid indicator id, person id, and term id to create a unique Id, which can be similarly done on other data feeds.

2. CSV files should not have headers
3. CSV columns should be in order matching field position in documentation
4. CSV files should use UTF-8 (or raw ASCII) encoding without BOM
5. Empty fields in CSV files may be set to "" or null
6. CSV files should be properly formatted to take into consideration the possibility of commas or double quotes in a data element.
7. Additional columns may be added to the end of a CSV file and used for logic or calculations for other fields
8. Additional or separate files may be used and combined into one feed using Accelerate queries

Connect Method 2: Direct DB - Views

Using a direct database connection with Connect2 hosted on the institution's Connect Server.

1. Ids must be unique

   Note: Many times, this Id is a combination of values imported on other columns that together create a unique identifier. For example, on the Person Financial Aid Indicator data feed, you may use a combination of the financial aid indicator id, person id, and term id to create a unique Id, which can be similarly done on other data feeds.

2. Field names are case sensitive
3. Results of SQL View must match SIS Import definitions

Connect Method 2: Direct DB - Queries in Accelerate

Using a direct database connection with Connect2 hosted on the institution's Connect Server.

1. Ids must be unique

   Note: Many times, this Id is a combination of values imported on other columns that together create a unique identifier. For example, on the Person Financial Aid Indicator data feed, you may use a combination of the financial aid indicator id, person id, and term id to create a unique Id, which can be similarly done on other data feeds.

2. Field names are case sensitive
3. Results of SQL Query must match SIS Import definitions