Week zero: what to capture before they go
The notice period is the most valuable and most wasted asset in this situation. Most handovers become a document nobody reads, because they are written as a list of systems rather than a record of decisions.
Ask for the things that are not discoverable by looking at the account. Anyone can see that a workflow exists; nobody can see why it was built that way, or which of the three similar saved searches is the one finance actually uses.
- The month-end close sequence, including which steps are manual and what to do when a period will not close.
- Which saved searches feed which reports and dashboards — and which are abandoned.
- Every integration: what it connects, who owns the other end, and how it fails.
- Scheduled and Map/Reduce scripts: what runs, when, and what happens if it does not.
- Any known problem currently being worked around rather than fixed.
- Credentials, tokens, and their renewal dates. Especially renewal dates.
Week one: access and inventory
Two things matter immediately. First, revoke the departing administrator's access on their last day and confirm no integration was authenticating as their user — this is a common and unpleasant discovery.
Second, produce a written inventory of what exists. Scripts, workflows, custom records, integrations, scheduled jobs, and anything currently erroring. You cannot triage what you have not listed.
- Audit who holds Administrator role and reduce it to the minimum.
- Check whether any integration or script runs under a personal user account.
- List every deployed script and its last execution status.
- Read the error logs — assume something has been failing quietly.
Week two: find what is already broken
In most accounts, something has been failing for a while and the departing administrator was either fixing it manually each month or had not noticed either. This is the week to find it.
Start with the integrations and scheduled jobs, because those fail silently. Then reconcile a recent period against the source systems and see whether the numbers tie.
Weeks three and four: the queue and the close
By now the request backlog has grown. Work it in priority order and publish that order, because the fastest way to lose the finance team's confidence is for their requests to disappear into silence.
Then get through one close. Nothing tells you more about the true state of an account than completing a month-end in it.
Deciding what to hire back
The instinct is to reopen the same req. That is worth pausing on, because you now have an opportunity you did not have before: real data about how much work the role actually generates.
Log the hours through these thirty days. If the honest figure is twenty hours a month, a full-time salary buys capacity you will not use — and a fractional administrator on a monthly retainer covers the same ground while you decide. If the figure is closer to a hundred, you have confirmed the full-time role is real and you can recruit for it with confidence rather than assumption.
- Measure the actual monthly hours before writing the job description.
- Separate recurring administration from the project backlog that piled up.
- Consider covering fractionally while recruiting, so the queue and the close continue.
- Whatever you choose, require documentation as a deliverable this time.
Shahin Zakizadeh
Founder & Principal NetSuite Consultant at SZnetsuite — SuiteScript development, automation, integrations, and billing for growing NetSuite teams.
About the practiceRelated services
Keep reading
What Does a NetSuite Administrator Actually Do? A Month in the Role
The weekly, monthly, quarterly, and twice-yearly work that keeps a NetSuite account healthy — and what starts breaking when nobody owns it.
Read articlePerformanceNetSuite Performance Optimization Checklist
A practical checklist for finding bottlenecks in searches, scripts, workflows, integrations, and user experience.
Read article