When bank feeds go down, the work does not stop, it just goes quiet. Deadlines still stand, clients still expect answers, and BAS does not slow down for a technical issue. What changes is how the work gets done. The feed drops and the accountant quietly switches to memory, judgment, and manual checks to keep the file moving, and none of that effort shows up in a dashboard anywhere. That invisible workaround is what actually finishes the file during an outage, and it is also where the risk hides, because software behaves as though it never happened.
The unofficial workflow accountants fall back on
Almost every firm follows the same pattern when a feed fails, even the ones that have never written it down. A CSV gets pulled straight from the bank. Transactions get scanned by eye rather than checked systematically. Familiar merchants get trusted faster, unfamiliar ones get parked for a later review that may or may not happen. This is not carelessness, it is pragmatism. The work has to continue and the tools do not offer anything better in the moment, so experience fills the gap.
The catch is that this workflow runs on judgment the software cannot see. An experienced accountant is holding context in their head instead of relying on consistent data, second-guessing transactions they would normally trust, and spending more time confirming that something has not changed than verifying that it is actually right. That is why a file worked during an outage feels so much heavier than a normal reconciliation, even when the transaction count is identical. The load is not in the numbers. It is in everything the accountant is quietly tracking that the system used to track for them.
Where the risk quietly enters the file
The danger during an outage is not the obvious mistake. It is the series of small compromises made to protect momentum. A description gets accepted as close enough. GST gets assumed from a familiar pattern rather than confirmed. A duplicate check gets done in someone's head instead of systematically. Each of those decisions is reasonable on its own, and each one is invisible afterwards. Together they quietly weaken confidence in the file long before BAS preparation ever begins.
This risk is amplified by the fact that most accounting software assumes a clean, continuous feed and simply does not adapt when that assumption breaks. It offers instructions, not a workflow. It tells you what to do, not how to cope when the data is arriving in fragments under a deadline. So there is a real gap between the official guidance and what actually happens at the desk. An experienced accountant feels that gap immediately and works around it on instinct. A junior often has no idea how much judgment is being applied behind the scenes, which is exactly how outage-era compromises get inherited by whoever reviews the file next.
How experienced firms protect confidence during outages
Firms that consistently avoid downstream problems do one thing differently. They separate the survival work from the final posting. Rather than repairing data directly in the ledger, they create a short review layer where the manual data is checked, normalised, and confirmed in bulk before it ever becomes part of the official record. That gives the accountant's judgment somewhere to live and be seen, instead of letting it silently harden into the file where nobody can question it later.
When bank feeds are down, accountants do not stop working, they work differently, and the whole file depends on judgment that never appears in a system. Recognising that is the key to a safer workflow. The firms that come through an outage without downstream BAS anxiety are not the ones with fewer disruptions. They are the ones who give that hidden judgment a visible, deliberate place to be confirmed, so a temporary disruption never quietly becomes permanent uncertainty.
