Skip to main content
Access review campaigns show reviewers a recommendation for each access grant, backed by usage and HR signals. The employeeAccessInsights download context exports that same underlying data as a CSV for every employee and application in your workspace - without creating a campaign, assigning reviewers, or waiting for a review cycle to close. This export is read-only. Requesting it does not create a campaign, notify employees or reviewers, or change any access.

When to use this export

  • Scope a campaign before you launch it. See how many grants would come back with a low-confidence recommendation, and which applications drive that count, so you can scope reviewers onto the access that actually needs a human decision.
  • Audit access outside a review cycle. Pull point-in-time evidence about who holds access to what and how they use it, on your own schedule.
  • Analyze insights in your own tools. Load the CSV into a warehouse, BI tool, or spreadsheet and segment by department, manager, or application.
  • Validate the signals against what you already know. Compare a signal to ground truth you hold today before you rely on it in a live campaign.
  • Integrate with your own access review workflow. Use the same signals and recommendations Oleria uses in its campaigns, but in your own system.

What each row represents

Each row is one employee’s access to one application, along with the insight signals Oleria computed for that pairing and the resulting recommendation. Columns are grouped by prefix:

Request the export

The employeeAccessInsights context uses the standard asynchronous download flow. See Working with Downloads for the full lifecycle, and the filterBy schema for the operators valid on each column type.
1

Get an access token

The API uses the OAuth 2.0 Client Credentials grant. Exchange your client_id and client_secret for a short-lived token, then send it as a Bearer credential on every request. See API Overview for authentication and for the base URL that replaces YOUR_TENANT below.
2

Create the download request

POST /v1/downloads with context set to employeeAccessInsights. You get back a DownloadRequest with a UUID id and status: "accepted".
3

Poll until it completes

GET /v1/downloads/{id} until status is completed and the response carries a presigned url. Back off between polls - a full-workspace export covers every employee and application pairing, so it can take longer than a narrowly filtered one.
4

Fetch the CSV

GET the presigned url directly. No Oleria authentication is required on the presigned URL. It expires after 5 minutes by default; set downloadUrlTtlMinutes on the create request if your pipeline needs longer.
Set fileFormat to jsonl on the create request if you would rather load newline-delimited JSON than CSV. The columns are identical either way.

How to read the ratings

Every signal, and the overall recommendation, uses the same scale:
A rating always describes confidence in keeping the access, never the size of the underlying number. A dormant_days_rating of HIGH means the identity was active recently, not that it has been dormant a long time.

The four signals

Oleria rolls these four signals into the single recommendation column. That value is the same recommendation reviewers see in an access review campaign, so a grant that reads LOW here would arrive in a reviewer’s queue flagged as low confidence.

How peer groups are defined

The peer group signal asks how unusual this access is among comparable employees, not how much the employee uses it. An application that every peer holds looks like standard access for the role. An application only this employee holds is an outlier worth a reviewer’s attention, however actively it is used. The peer group is a workspace-level setting. Go to Governance -> Access Reviews -> Settings -> Peer Group and choose one of:
  • Department
  • Job Title
  • First-level Manager
  • Second-level Manager
Changing this setting changes peer_group_percentile and peer_group_rating in later exports. If you compare two exports taken at different times, confirm the setting matched for both.

Column reference

Employee and identity

The employee’s HR record and the identity provider account that holds the access. The two can disagree - subject_email comes from the HR system, subject_identity_email from the identity provider.

Application

The application the employee has access to.

HR change signal

The old and new columns are populated only for the attribute that changed. If hr_change_job_title_change is false, hr_change_old_job_title and hr_change_new_job_title are blank.

Dormancy signal

AT_LEAST_ACTIVITY_WINDOW_DAYS is a floor, not an exact count. For the equivalent figure in the Oleria interface, see Dormancy thresholds, where the same case is displayed with a + suffix.

Login frequency signal

Peer group signal

Recommendation

Any column can be blank when the underlying data is not available or does not apply to that row. A blank rating column is written as UNAVAILABLE; a blank value column is written as an empty string.

Contact us

For questions about this export, contact us at support@oleria.com.