What does CAS client onboarding involve before the first monthly close?
Agreeing the firm’s chart-of-accounts conventions, file naming, checklists and reviewer preferences; fixing the scope and review points before work starts; reconciling or listing the opening position; and running one client end to end before volume is added. Treatment decisions and sign-off stay with the firm.
Most CAS engagements are judged on the first close, and most first closes are judged unfairly. The period is doing two jobs at once: producing a month, and discovering every convention the firm never had to write down because the same person always did the work. Those are different activities, and treating the first as though it were only the former is the usual reason a handover feels slower after it happened than before.
This covers what to settle before the first close runs. It sits under CAS production support and assumes the engagement scope itself is already decided — the CAS guide covers that part.
Conventions before transactions
The first thing to transfer is not data, it is convention. A firm running recurring closes has accumulated dozens of small standing decisions: which account a particular recurring cost is coded to, how a memo is written so the reviewer can scan it, what a file is named, which schedules the reviewer expects to see and in what order. None of it is wrong and almost none of it is documented.
Structured discovery means writing those down before production starts: chart-of-accounts conventions, file naming, the firm’s own checklists, deadlines, reviewer preferences, and the exceptions that recur every month. An hour spent on this is the cheapest hour in the engagement, because every item not captured becomes a correction later — and a correction that arrives in month three has already been made wrongly twice.
Scope and review points, agreed before work starts
Scope here means more than the list of accounts. It means which entities and periods are included, what the output is, which steps the firm reviews and which it accepts, and who answers a question when the records conflict with each other. Agreeing the review points in advance matters more than it sounds: a preparer who knows which three things the reviewer looks at first will build the file so those three things are visible.
This is also where the delegation line is drawn, and it does not move afterwards. Production work — posting, reconciling, assembling the package — can be handed over. Choosing an accounting treatment, interpreting the result for the client and signing off stay with the firm’s licensed staff.
The opening position is the part that goes wrong
A recurring close starts from a balance, and the balance arrives from somebody else. What a new engagement needs at the handover date is a trial balance, prior-period statements, bank and card statements covering the opening period, the current chart of accounts, open AP and AR detail, payroll reports, loan and lease schedules, and any recurring journal entries the previous preparer was posting.
Where the records support a reconciliation, reconcile. Where they do not, list the difference with its amount and account and leave it visible. The temptation at this point is to clear an unexplained opening difference into a suspense or equity line so the balance sheet ties — and it is worth naming plainly, because a first close that ties perfectly and quietly is harder to trust than one that ties with three named exceptions. The firm then decides whether the difference needs separate cleanup before a normal monthly cycle can run.
Onboarding and cleanup are often needed together and should be scoped apart. Cleanup corrects a prior period and has an end point; onboarding sets up how the recurring work will run. Combining them into one undefined block of work is how both end up with no completion criteria.
One client, end to end, before volume
Run a single client through a complete cycle before adding more. One full close surfaces the undocumented conventions, the records that do not arrive, and the reviewer’s actual preferences at a scale where fixing them costs almost nothing. Onboarding five clients in parallel does not find those problems five times faster — it reproduces one undiscovered problem five times and makes the cause harder to isolate.
The feedback loop is the deliverable
The corrections from the first two closes are the most valuable output of onboarding, and the most commonly discarded. Each one is a convention that existed only in someone’s head. If they live in a message thread they are gone by the next cycle; if they are written into the workflow documentation, the engagement gets cheaper to review every month instead of staying the same.
That is the actual test of whether onboarding worked. Not whether the first close was fast — it will not be — but whether the second one needed fewer of the same corrections than the first. For what the recurring cycle looks like once the conventions are settled, see month-end close and accounting support and recurring bookkeeping. Where the incoming records need correcting first, our bookkeeping outsourcing guide covers the cleanup sequence.