This tool does two jobs for the Learn365 → Absorb move. First, it lets one person review every Learn365 course and record what should happen to it in Absorb. Second, it copies learners' training history (completions and, if you choose, open enrollments) into Absorb and checks that the copy is right. Work through the steps in order. Nothing is written to Absorb until step 5, and even then a dry run comes first.
Step 1Connect to Learn365
Open Settings. Paste a Read Only Learn365 API key (Admin Center → Global Settings → API key management). Leave the base URL as https://api.365.systems/odata/v2 unless Zensai tells you otherwise.
Click Test connection. The probe lists the field names Learn365 actually returns. If a value on the review page comes back blank later, the field name goes in the Field mapping box on this page.
Step 2Sync
Click Sync from Learn365. Courses, learners and enrollments are pulled into the local database. Re-run it whenever Learn365 changes; the review decisions you have already made are kept.
Optional but recommended: add the Absorb sandbox credentials in Settings first, then Pull Absorb courses & departments on the Migrate page, so the review page can suggest Merge targets.
Step 3Review each course
Pick a course from the list. The page shows how much it is used, when it was last completed, a preview of its content, a possible Absorb match, and why a suggestion is made. Then choose one decision:
1 Rebuild — recreate properly in Absorb. 2 Lift as-is — copy the files over unchanged. 3 Merge — history goes into an Absorb course that already exists; enter its Absorb course ID. 4 Retire — no Absorb course; history is not carried. 5 Needs SME — park it for a subject-matter expert.
Keys: J/K next/previous course, 1–5 decision, R mark reviewed, L hide the list, Ctrl+Enter save and move on. Use Bulk to retire many stale courses at once.
Step 4Set the migration decisions
Open Migrate. Keep the target on sandbox until the trial passes. Then set:
D1 — how far back to carry completions (blank = everything). D2 — whether people who have left get an inactive Absorb account so their history has a home. Active learner with no Absorb account — leave on Hold; HRIS/SSO creates the account and the record loads on the next delta. D3 — whether unfinished enrollments are carried or left for Absorb's rules to re-assign.
Click Test Absorb connection, then Save.
Step 5Plan, dry run, load
Build plan turns your review decisions into a course map and lists anything that would block a load (no email, unknown status, completion with no date, unmapped department). Fix or accept each item.
Dry run writes nothing and produces load_plan.csv showing exactly what a real load would do, learner by learner. Read it. Only then Commit. The database is backed up first, users are matched by email and never altered, and a production load asks you to type production.
Step 6Verify, then repeat daily
Verify reads every loaded record back from Absorb and reports OK, missing user, missing enrollment, wrong status or wrong date, with a CSV of the exceptions.
Loads are safe to re-run: records already loaded are skipped, so a second commit writes nothing. During hypercare turn on Daily delta; it extracts new Learn365 activity, loads it and verifies it each night. Reset target forgets what was loaded to a sandbox so the trial can start clean.
Nevvon (user data only). Nevvon has no API, so the Nevvon page works from a file. In the Nevvon panel run Reports → Completion Report (aggregate per employee, all programs and locations, full date range), download it from the Notification Center, and upload it. The tool detects the columns; fix any it guessed wrong, choose whether rows are programs, modules or lessons, and Import. Then map each Nevvon program to the Absorb course that should hold its history, or skip it. From there Nevvon records travel through the same Plan → Dry run → Commit → Verify steps as Learn365, marked source = nevvon. Re-upload a newer report any time; rows already imported are ignored.
Content (Lift as-is packages). Absorb's API cannot upload a SCORM or HTML package, so the Content page drives the Absorb admin portal with a browser worker instead. Store each Lift-as-is course's zip (fetched from Learn365 or uploaded), fill the selector file from one recorded manual upload on the sandbox, sign in once on your own PC with the login helper and upload the session, then queue a dry run. When dry runs pass, queue the real upload; each finished course gets its Absorb course ID written back so the history load can target it. Nothing on this page can run until an Absorb sandbox exists.
Sandbox trial checklist. Before touching production, prove on the sandbox that: the original completion date is kept; no emails or certificates go out; loading history onto a learner who already holds the course leaves their live enrollment untouched; and account creation succeeds under the portal's password policy. If any of these fail, stop and adjust the Absorb route overrides or settings before going further.
Settings
Learn365
Review rules
Absorb — sandbox (used for merge suggestions and for trial loads)
Absorb — production (used only when Migrate target is set to production)
Field mapping (advanced)
The app tries several likely field names for each value. If the probe below shows different names in your tenant, override them here as JSON, e.g. {"course.title":"Name","enr.completed":"FinishedOn"}.
Probe result
Content · move "Lift as-is" packages into Absorb
Absorb's API cannot upload a learning object, so a browser worker (Playwright) does the admin-portal upload for each course you marked Lift as-is, then writes the new Absorb course ID back so the history load can use it. Until an Absorb sandbox exists, steps 1 and 2 can be prepared; 3 and 4 wait.
One zip per Lift-as-is course. Fetch asks Learn365 for the package (works only if the API exposes a download URL for that course); otherwise download it from Learn365 yourself and Upload it. The tool reads imsmanifest.xml to confirm what kind of package it is.
2 · Absorb admin selectors
This JSON describes the clicks the worker makes in the Absorb admin portal. It is a template with TODO placeholders because no Absorb tenant has been seen yet. On day one of sandbox access: record one manual SCORM upload with python -m playwright codegen https://TENANT.myabsorb.com/admin, paste the selectors here, remove every TODO, and set admin_url. Placeholders {title}{description}{external_id}{category}{package} are filled per course. stop_if_dry_run marks where a dry run stops (fill everything, never Save).
3 · Absorb session
How it worksSign in once on your PC, upload the sessionMFA cannot be done headlessly on the server. Download absorb_login_helper.py, run it on your computer, sign in to the Absorb admin (MFA included), press Enter, and upload the absorb_state.json it writes. The worker reuses those cookies. Treat the file as a password; remove it here when the migration window closes.
Server browserchecking…
4 · Run
Runs one course at a time. A dry run fills every form and uploads the package but stops before Save, and leaves a screenshot. Courses already holding an Absorb course ID are skipped, so re-running is safe. Every job keeps a Playwright trace (trace.zip) and a final or failure screenshot under /var/lib/learn365/shots/.
Nevvon → Absorb · user data only
Nevvon has no API. Export a Completion Report from the Nevvon panel (Reports → Completion Report → Notification Center → Download), upload it here, map its programs to Absorb courses, then load it from the Migrate page with everything else.
1 · Upload a Completion Report (CSV or XLSX)
Staged fileCheck the column mapping below. Email (or Employee ID), the program/module/lesson name, and a completion date are required. Everything else is optional.
Set each program to Map with the Absorb course ID that should receive the history, or Skip if it should not be carried. Pull Absorb courses on the Migrate page first to get title-match suggestions. Saved decisions appear in the Migrate course map and load with the next plan.
3 · Load
Nothing else to configure. Go to Migrate → Build plan → Dry run. Nevvon records appear alongside Learn365 records with source = nevvon in load_plan.csv. Learners with no email in the report are flagged NO_EMAIL and never loaded; leavers and active-learner holds follow the same D2 rules.
Removes imported Nevvon programs, records and decisions. Learn365 data is untouched.