One ledger with property classes, or a separate file per entity. The books I read split about even

Question

I've gone through the books of four operators this year while deciding where to put money, and the split was two and two.

Two of them ran everything in a single accounting file with each property tagged as a class or a location, one bank feed set, one reconciliation cycle. The property-level P&L came out fine and the bookkeeping bill was one bill. When I asked how they kept the entities straight, the answer was intercompany accounts and a lot of discipline.

The other two ran a separate file per LLC. Four subscriptions, four closes, four sets of reconciliations. Their argument was that each entity files its own return and holds its own bank account, so the ledger boundary should match the return boundary, and that when a partner buys into one entity you hand over one file and nothing else moves. One of them also said his attorney pushed for it on separation grounds, and how much weight that carries depends on your state and your operating agreements, so that's a question for a lawyer rather than for me.

The cost difference is real. Single file quotes I got ran roughly $300 to $400 a month for nine doors. Per-entity quotes ran $150 to $200 per entity per month, so three entities landed higher even though the transaction count was identical.

What I can't settle is whether the single-file operators are carrying a hidden risk that only shows up when something goes wrong, or whether the per-entity operators are paying a premium for tidiness.

mitchell_ray4 · Accepted Answer

Coming at this from outside accounting, the rule I ended up using is simple. A separate file wherever there's a separate tax return and a separate bank account. Classes for everything below that line.

The reason is that the return is the thing that has to be defensible later, and every reconciliation you do inside one file is proof that the cash in that entity matches the ledger for that entity. When you run three entities through one file, you're reconciling the bank accounts individually anyway, so you've done most of the work already, and what you've saved is the software seat plus some setup time. What you've given up is that a misposted transfer between entities looks like a normal internal move instead of an obvious out-of-balance.

@tally, to your question, the preparer can filter by class, but somebody has to confirm that every transaction carries a class, and in every one-file setup I've seen there's an unclassified bucket that has to be cleaned up before anyone can use the numbers. That cleanup is billable time and it usually lands in February.

On @flint's $40k, in a per-entity setup you'd have booked a due-to and a due-from in two files, and either one of them not reconciling at month end tells you immediately.

Whether the entity separation matters for liability purposes depends on your state and how the entities are papered, and that's an attorney question, so I don't lean on it as the reason. The reason I lean on is that the ledger boundary matching the return boundary makes errors visible.

stephen_ok9 · Answer

Adding the numbers I left out. The single-file operator with nine doors was paying $340 a month all in, one software seat at $90 and the rest labor. The three-entity operator with eight doors was at $185 per entity, so $555, plus three seats. Same transaction volume within about 15 percent. That's a $215 a month spread, $2,580 a year, for a structural choice that has nothing to do with how many receipts anyone is coding.

The part I keep circling is that the cheaper setup only stayed clean because one person cared a lot. When he took two months off, the intercompany balances drifted and it took a bookkeeper nine hours to unwind.

joe_delgado5 · Answer

One file, classes, and I'm not paying for four subscriptions to feel organized.

I still got hit exactly the way you'd expect. Moved $40k from the holding entity to the flip entity to cover a draw, my bookkeeper coded it as owner contribution in one direction and never booked the other side. Two quarters later the two intercompany accounts were off by $40k in opposite directions and it took a full afternoon to find because it all lived in the same file and nothing looked wrong at the top level.

In separate files that error screams at you. In one file it hides.

aliyah_b · Answer

Question from the cheap seats since I'm still on my first deal. If everything is in one file with classes, what actually goes to the preparer at tax time? Does he just filter the P&L by class and call that entity one, or does someone have to rebuild it? That step is where I'd expect the savings to evaporate.

tonyv88 · Answer

There's a category of this that isn't a preference at all. If you're holding other people's money, tenant security deposits or owner funds you manage, most states have rules about how those accounts are kept and reconciled, and the requirements differ by state, so check your own before you decide anything. In practice that has meant a separate ledger for the trust account no matter how the rest is arranged.

For your own properties, I've run both. One file with classes is fine right up until you sell one entity, and then you spend a week carving out history.

tylerbrooks9 · Answer

One file per return. It's the version I can explain to myself in a sentence, and I've never regretted paying a bit more for that.

marcus_hale41 · Answer

As someone reading these books from outside, the file layout is close to irrelevant to me. What I ask for is the bank statement for the period, the reconciliation report showing zero uncleared items older than 60 days, and whether the intercompany accounts net to zero across entities. An operator running one file who can hand me those three things in a day is in better shape than one running four files who can't.

Where it does matter is the carve-out @trellis mentioned. Twice now I've looked at a single-entity acquisition where the historical financials had to be reconstructed out of a combined file, and both times the numbers I got were adjusted, not reconciled. That costs on price.

angela_burke · Answer

Pricing this from the other side of the desk. Closing four files with 60 transactions each takes my team meaningfully longer than one file with 240, because the fixed cost per close is real. Open the file, pull the feeds, reconcile every account, run the reports, review, deliver. Call it 45 to 60 minutes of overhead per file per month before anyone codes a single line.

So four entities at 60 transactions is roughly three extra hours a month of pure setup and teardown against the combined version. At $65 an hour that's your $200 spread, @meridian, and it's not padding. It's the actual work.

What I tell people is that if you're going to run per-entity, keep the entity count down to what the structure actually needs. I've seen five LLCs where two would have done the same job, and the owner was paying $9,000 a year in bookkeeping to maintain a diagram.

One ledger with property classes, or a separate file per entity. The books I read split about even

8 replies