Import field reference

Check a file against this list before you import employee data. These are the columns every import lands in. A file laid out another way is mapped onto them for you to review before anything is written, as Importing employee data describes. Whatever the layout, every employee needs a work email address.

How the file is read

• The first row is the header. Column names are matched without regard to case or spacing, so Work Email matches work_email.
• In a file already laid out in these columns, they can come in any order, and every column except work_email can be left out.
• An empty cell in an optional column is allowed.
• A file can hold up to 50,000 rows.
• Each row is matched to an existing employee by work_email. A row whose address isn’t in your employee data yet adds a new employee.
• A column that isn’t in the table below is an extra column. You decide on the review what to do with it: make it a custom attribute, fill one you already have, or leave it out.

Columns

Column	Required	What it holds
work_email	Always	The employee’s work email address. It must be a valid address, and each one can appear only once in the file.
given_name	For a new employee	First name.
family_name	For a new employee	Last name.
title	No	Job title.
department	No	The department’s name. A department that doesn’t exist yet is created.
manager_email	No	The work email of the employee’s manager, who can be another row in the same file.
dotted_line_manager_emails	No	The work emails of any dotted-line managers, separated by ;.
phone	No	Phone number.
location	No	Office or city.
employment_status	No	active or terminated.
start_date	No	The date they started, as YYYY-MM-DD.
birthday	No	YYYY-MM-DD, or MM-DD without the year.
pronouns	No	Pronouns.
base_salary	No	Annual base pay in whole currency units, not cents, such as 85000. Commas, spaces and $ are ignored. It can’t be negative or above 100,000,000.
currency	No	The pay currency as a three-letter code, such as USD or EUR. USD when left out.
fte	No	Full-time equivalent, more than 0 and up to 1, such as 0.5. 1 when left out.
job_id	No	Your own job code.
grade_id	No	The pay grade or level code. This is what matches an employee to a pay band.
performance_rating	No	Any value from your own performance process, or blank for not rated. See Performance reviews.
photo_url	No	A link to the employee’s photo.

When a row is refused

A value in the wrong format is reported against its row, with what was wrong and the format to use, such as Invalid start_date: 03/01/2024 (use YYYY-MM-DD). A new employee without both given_name and family_name is reported the same way.

Reviewed file layout

Download reviewed file, on the review’s Review and approve step, gives you reviewed.csv: the file the import will run, as Reviewing an import describes. It holds one row for each row that will import, and leaves out the rows that can’t, and the rows you skipped. Its columns are, in this order:

1. The columns above that your file carries, in the order of the table in Columns, with photo_url last. A column your file doesn’t carry isn’t added.
2. Then each extra column the import writes, in the order your file has them, each under the name it imports as: the name of the existing attribute it fills, or for a new attribute the name you gave it on Extra columns. The header is the attribute’s name as it will read in Mitratech Comp, not the header your file had.

A column you chose Don’t import for isn’t in the file. Neither is any column your role can’t see: without access to pay, base_salary is left out, and a custom attribute’s column is left out unless your role can see that attribute. A new attribute starts visible to administrators only, so only they get its column. See Custom attributes.

Review file layout

Select Import review results, on the Performance Reviews page, to load a review file instead of the layout above. It carries two required columns and as many question columns as the review needs.

Column	Required	What it holds
work_email	Always	The rated employee’s work email address. It must already belong to someone in your employee data; a review-scoped import does not create anyone new.
performance_rating	Always	Your own rating value for this employee. Blank leaves them unrated.
any other column	No	One of the review’s questions, named by the column header, with each row’s cell as that employee’s answer. Keep an answer under 10,000 characters.

See Performance reviews for what a review’s questions and answers look like once imported.

Guidance override file layout

Select Import overrides in a cycle’s 3. Guidance section to load a guidance override file instead of the layout above. It sets or clears guidance overrides for that cycle only, and it changes nothing else about an employee. It must be a .csv file laid out in these columns; no other layout is mapped for it.

Column	Required	What it holds
work_email	Always	The work email of an employee the cycle plans. Each one can appear only once in the file.
guidance_override_low_pct	In a cycle that plans ranges	The low end of the employee’s guidance range for this cycle, as a percentage from 0 to 100, such as 4 or 4%. Leave both range columns blank to clear the employee’s override.
guidance_override_high_pct	In a cycle that plans ranges	The high end of the same range, as a percentage from 0 to 100, not below the low end. Leave it blank with the low end to clear the override.
guidance_override_pct	On any other cycle (Flat, or one that doesn’t plan ranges)	The employee’s guidance for this cycle as a single percentage from 0 to 100, such as 6 or 6%. Leave it blank to clear the employee’s override.
guidance_override_note	No	A note on why, up to 500 characters. Leave it blank to clear the note. Without this column, an override you set again keeps its note.

On a cycle that plans ranges, the file must carry guidance_override_low_pct and guidance_override_high_pct, and guidance_override_pct is not read. On any other cycle (Flat, or one that doesn’t plan ranges) it is the other way round. A row with only one end of its range, or a low above its high, is skipped with its reason. A range that passes a hard bound of the merit guardrail is imported with a warning, not skipped.

A file without work_email or the range or percentage column for its cycle is refused whole, naming the missing column. An employee the file does not name keeps the guidance they have. See Building a merit matrix or pay guidance for what is checked before you approve.