Flux & Variance
The accounts you can flux fastest are the ones you booked yourself
Flux difficulty doesn't track account size or dollar movement. It tracks how well you understand how the entry got there — which is worth knowing, because it changes what you can prepare for.
There are accounts on your schedule that take ten minutes and accounts that take an hour, and the split usually isn't about the size of the variance.
It's about whether you know what's inside the entry.
If you prepared the journal yourself, you already know what the moving pieces are, which ones vary month to month, and which direction each one pushes the balance. The explanation is mostly transcription. If someone else prepared it, you're reconstructing all of that from the ledger before you can write a sentence — and for an entry with several components, that reconstruction is most of the work.
This is more obvious the first time than it is later. An account with a few moving parts can be genuinely hard the first month you touch it and routine by the third, with nothing about the account having changed. What changed is you.
Why this is worth naming
Because it means flux readiness isn't a property of the account. It's a property of the relationship between the preparer and the entry, and that has two consequences worth acting on.
The first consequence is about who does what. Where preparation and explanation are split across people, there's a recurring cost that nobody books: the explainer re-deriving what the preparer already knew. That cost is invisible because it looks like the account being hard.
The second is about timing, and it's the more immediately useful one.
You don't have to wait for close to finish
Most flux gets treated as a single block of work that starts once everything is closed. That's not actually required. An account can be explained as soon as everything that feeds it has landed, not when the whole close is done.
Which means there's a sequence hiding inside close that most schedules don't exploit. Once the activity feeding a set of accounts is final, those accounts are ready — regardless of what's still open elsewhere. The accounts that depend on the last entries of the period are genuinely blocked. Plenty of others aren't, and they get treated as though they are.
The practical version: identify which accounts become explainable at which point in your close, rather than treating flux as one task at the end. Some accounts are ready on day two. Waiting until day five to write them up doesn't make them any easier, it just compresses everything into the window where you have the least time and are most likely to rush.
The failure mode when you do rush
An account with several moving components, explained under time pressure, produces a specific and predictable error: you get the direction wrong on one of them.
You know all the pieces are in there. You're moving quickly because the deadline is real. And you write that one component increased the balance when it actually reduced it — not because you don't understand the account, but because you didn't stop to check the sign on each piece individually before writing the sentence.
This is worth knowing about because it's the failure that reads fine. The explanation names real drivers, uses specific language, and sounds like someone who knows the account. It's just pointing the wrong way on one component. A reviewer skimming at the end of close will very often accept it, which is exactly why the directional check is worth running mechanically rather than by eye.
Where the manual time actually goes
When an account like this is slow, the slowness is usually structural rather than conceptual.
The pattern that eats time: the current period's detail lives in one place, the prior period's lives in another, and comparing them means pivoting each separately and then matching them up by hand. Nothing about that is difficult. It's just several manual steps repeated every month, on an account you already understand, producing a comparison you already know how to interpret.
That's the part worth attacking, and it's attackable precisely because it requires no judgment. Pivoting detail by counterparty, pulling the equivalent figures from a prior period, and lining them up is compilation. Deciding what the resulting movement means is not.
Accounts where this already works well tend to be the ones where the supporting process was built first — a reconciliation that completes before the flux runs, in a consistent format, so the comparison is a step rather than a project. Accounts that sit outside that structure stay manual, not because they're harder, but because nobody built the structure around them.
The thing this points at
Most of the effort in a close is spread across recurring entries and reconciliations that look different every month but are structurally identical every month.
If you prepare an entry regularly, you already know its shape: which components appear, which vary, where the supporting detail lives, what the comparison against last period requires. That knowledge currently lives in your head and gets re-executed manually every period.
Designing the process around your actual recurring entries — rather than treating each close as a fresh problem — is where the time is. Not in explaining better. In not rebuilding the same comparison from scratch twelve times a year.
Most of this surfaces again at flux review. The free flux template flags which balances moved enough to explain. The CloseOps Flux & Variance System ($79) starts from your trial balance instead: confirm each account's classification and it builds the income statement and balance sheet flux statements, then ranks what to investigate.
There's a free tool for the sequencing part — tick off what's closed, see what's ready to explain. Related: why income statement and balance sheet flux need different data, what actually goes into an accounting estimate, why some accounts can be explained from the detail and others can't, and what happens when a late entry moves an account after commentary is written.
Found something wrong or missing? Tell me here — anonymous, thirty seconds.