The River Problem
Digital money organized into perfect channels, just floating peacefully into the clean books of the modern accountant, with their perfect rules and booleans and notifications. This is the romanticized version that accounting tech has been chasing. Like a profitable, lazy, river, transactions flow in continuously. Bank feeds, payment processors, e-commerce integrations, payroll syncs, and anyone that can spell API can sync to the general ledger, which sits downstream, collecting whatever arrives, in whatever shape it's in. The pitch is efficiency.
If you live in the bank feed world and have ever spent the last day of a close trying to figure out why operating cash doesn’t reconcile, the problem almost certainly did not start at month-end. It started the moment you let an unreviewed data stream write directly to your GL.
The Architecture Problem
General ledgers are not designed to receive raw transactions. They were designed to receive reviewed, approved, summarized journal entries from subsidiary systems (subledgers) that had already been controlled at the source. Accounting principles. Old accounting systems assumed humans would validate activity before it became accounting.
Payroll posts one entry to the GL. AP posts one entry. The prepaid schedule runs its amortization, someone reviews it, and one entry posts. Every data stream is segmented. Every batch has a review point before it touches the GL. That is the architecture. It is not complicated. It is deliberate.
I remember the first time I encountered bank feeds. I was newly minted into my first industry job, quite a bit over my head as a manager of a bunch of things, many of which I had never done before. Despite spending a decade as an auditor, I was pretty fresh behind the wheel of a company's actual accounting function.
Quickbooks online was the shiny new vehicle, and bank feeds held the promise of making my life a cakewalk of monitoring transactions flow through a carefully crafted web of rules and automations. It all went swimmingly as I dove into the chaotic scramble of keeping a startup's books. All was well until I actually had to reconcile my first bank statement. Wow, did I have to check myself for imposter syndrome. No matter, I powered through. I wish I could say it was a hot stove moment, and I learned my lesson, but I chased the automation ghost for years before figuring out a happy medium. I've stepped back in time a bit, have a few more battle scars, and have taken a much deeper look at a concept that has largely been abandoned by the cool kids on the accounting podcasts. I am talking about batching.
The Benefits of Batching
Let's throw another term out: gatekeeping. No, not the toxic office version; the financial controller gates to the financials.
I am talking about control points. I am a Controller, after all. The Controller controls their financial domain by ensuring the integrity of data (read: cash) entering any part of the accounting cycle. So why did so many good Controller's let fintech into their domain without a checkpoint?
The promise of efficiency, of course. They bought what the software vendors were selling. And they were selling the user experience, hard.
It all looks so clean, you see. Professionals have become very aesthetic with our software. Maybe it was well intentioned, but ultimately, they make what we buy, and we buy what feels like it will make our lives easier. Do we really think all these snazzy interfaces we are creating are so useful? I think in hindsight, I did. But I've woken up. I am coming back around to batching.
What Batching Actually Means
Batching does not mean slowing down the process (materially). It simply means that transactions accumulate in a controlled environment before they enter the core accounting system of record, old General Ledger. A subledger, a staging file, a review queue, the naming is insignificant. What is significant is that they are reviewed as a group before they post. You can see the batch. You can check it. You can reject individual items. You have a record of what was reviewed, by whom, and when.
Compare that to a bank feed posting directly to the GL. By the time you review it, the entry is already in the books. It may be miscoded, duplicated, cross posted, copied over, posted to the wrong period, left an invoice hanging, or otherwise be flat out wrong for all the other reasons accounting is hard and machines won't replace us soon. And because hundreds of other entries have followed it downstream, you cannot as easily isolate what broke that one reconciliation or reverse it without creating more noise downstream.
The ultimate benefit of batching is that it forces good review habits. It forces a manned checkpoint into the general ledger. It is where professionals with the judgment to execute in these roles work our magic. Remove it from the workflow and you have a system that produces numbers without producing accountability, and that adds up to risk of misstatements and leakage.
Where the RoboBeancounters Take This
Given the recent trajectory of the software makers, it stands to reason that the feng shui software design philosophy continues to win out. But what if AI hammers the software industry as we currently know it? There is a growing cadre of voices starting to point out that when we can all custom design our own systems in a few late evenings, anyone smart enough to be doing so, will.
So what then becomes of all the software that has clogged up our lives? I care not to speculate, but I think speak for a lot of people that drastically want to simplify financial operations. I am telling you, my fellow overwhelmed accountants, the answer is going back to batching.
As many of us continually gain proficiency with AI tools, we are noticing that it is perfectly suited to the batching workflows of days past. How exactly we get there, is beyond the scope of this discussion, but what is powerful is the critical importance of inserting more review back into the workflow in a meaningful way. I think AI will be a great organizer of chaos in what has become a very digitally cluttered life and profession.
Joe Minich, CPA
Founder of Ridgeline Business Solutions. Joe works as an embedded financial operator and readies businesses for the accounting of the future. Ridgeline helps owner-led businesses in property, real estate, and skilled trades navigate growth, transition, and operational chaos.
