Flux & Variance

Balance sheet flux analysis: how to explain what moved

Balance sheet flux analysis compares an asset or liability balance to a prior period and explains why it sits where it does. The account carries forward, which changes what you pull, what you're asking, and what counts as an answer.

The mechanics are the same as any flux: compare, flag what moved materially, explain the drivers. What's different is that a balance sheet account doesn't reset. Its balance is the accumulation of everything that ever happened to it, and the current period's activity is only the most recent layer.

That single property drives most of what follows.

One month of detail is already the movement

For an income statement account, a month of activity is that month's balance — so comparing two months means pulling two months. For a balance sheet account, the prior period's effect is already sitting in the opening balance, which means the current month's activity is the change.

To explain what happened to a balance between 7/31 and 8/31, you run August's detail. That's it. Pulling July as well produces a larger dataset that answers a question you didn't ask.

When the comparison is against a fixed earlier point rather than the prior month — the prior fiscal year-end is the usual one — the same logic scales up. The movement you're explaining is everything between that anchor and now, so the pull runs the full span. That's a much bigger dataset, and it's where reading detail line by line stops being realistic and a category-level pivot becomes necessary — this free tool does the grouping if you'd rather paste the detail in than build the pivot by hand.

The full comparison with income statement pulls is here.

You're explaining a level, not an event

This is the part that changes what a good answer sounds like.

An income statement variance asks what happened during the period. A balance sheet variance asks whether the balance is where it should be given what the business has been doing — which is a question about accumulation, not activity.

So the explanation often points at a pattern over time rather than a single event: a build-up, a drawdown, a timing gap between when something is recorded and when it settles. "Higher activity in the period" is at least the right shape of answer for an expense account. For a balance that has been climbing steadily for four months, it explains nothing about why the balance is where it is.

The useful mental model is a rollforward. Opening balance, what was added, what was used or released, closing balance. Even when you aren't formally preparing one, framing the movement that way tends to surface the real driver faster than scanning transactions.

How the common accounts actually behave

Balance sheet accounts fall into recognizable patterns, and knowing which pattern an account follows is most of what makes the review fast.

Cash is close to self-explanatory. Deposits against payments, and the story is mostly visible in the detail. It's usually the quickest account on the schedule.
Receivables move on the gap between what was billed and what was collected. A balance growing faster than revenue is the standard thing a reviewer will ask about.
Inventory moves on the gap between what was built or bought and what was sold. Balances often build deliberately ahead of a busy period, which looks like unexplained growth if the comparison period sits on the other side of that build.
Payables accumulate and then clear on settlement. A large movement is frequently about payment timing rather than about spending more.
Accruals follow the same accumulate-then-release shape: booked in one period, used up when the actual cost lands (or reversed on day one of the next period, if booked that way). A balance that isn't moving the way you'd expect is worth a look, since an accrual that stopped moving entirely produces no variance flag at all.
Reserves and allowances are sized against whatever they're provisioning for, so their movement reflects a change in that underlying exposure or a change in the estimation method. What supports the method matters as much as the balance.

Seasonality shows up as balance size

Seasonality gets treated as an income statement concept — revenue concentrated in certain quarters, spend that follows it. Balance sheet accounts carry seasonal patterns too, and they appear as the size of a balance rather than as activity in a period.

An inventory balance built ahead of a busy season. A reserve legitimately larger at one point in the year than another because of what it's provisioning against. A payable that accumulates and then clears when settlement happens.

None of those are errors, and all of them look like unexplained movement if the comparison period sits on the other side of the pattern. Which of your balance sheet accounts have a seasonal shape is specific enough to each business that it mostly has to be learned rather than looked up — and it's a large part of what separates someone reviewing quickly from someone investigating every large movement from scratch.

What a size threshold misses here

Flagging on the size of a movement is standard and it works, but the balance sheet has a few failure modes it can't see.

An account that should have moved and didn't produces no variance, so nothing flags — a recurring accrual that silently stopped posting looks identical to one that's fine. A sign flip on a small balance can be genuinely wrong while staying well under any dollar threshold. And an account added during the period has no prior balance to compare against, so it may never enter the comparison at all.

Those are structural exceptions — they need a different comparison, not a lower threshold.

Which comparison to run

Worth confirming before you pull anything, because it determines the dataset and it isn't always the same one your internal review uses.

Month over month — against the prior month end. Current month's detail is the movement.
Against prior fiscal year-end — common for balance sheet accounts in reporting packages. Pull from that anchor forward.
Same month prior year — useful where the account has a seasonal shape, because it compares like against like.

A driver that explains this month against last month often says nothing about the balance against last year-end. They're different questions that happen to sit in adjacent columns of the same package.

If you work this in Excel: the free flux template flags what needs explaining and shows the residual your drivers don't cover. 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.

Found something wrong or missing? Tell me here — anonymous, thirty seconds.