Going digital in condominium management: the honest roadmap

You know the setup: a filing cabinet per building, one spreadsheet for the co-ownership shares, a second for the advance payments, a "resolutions" folder on the network drive, and the actual truth in the head of the colleague who has been there fourteen years. It works. It works remarkably well — until someone falls ill, until an owner asks to see the 2019 statement, or until you take over a building and discover the previous manager worked exactly the same way.

This article is about getting out of that. Not in one sweep, but in an order that survives an assembly season landing in the middle of it.

Why most attempts stall

The usual mistake is treating the move as a project with a start and an end. You buy software, block out a weekend, and on Monday everything is new. That fails, reliably, for three reasons.

The old data is worse than you think. Not wrong — incomplete. The co-ownership shares are right, but three units are missing the date the owner changed. Opening balances exist, but only as a balance, not as bookings. You find out during the import, and that is where the project stops.

Day-to-day work does not pause. While you migrate, defect reports, invoices and questions keep arriving. If the new system does not carry yet and the old one is half abandoned, you work in both for a while. That is the moment projects get cancelled.

There is no point at which it has paid off. Move everything at once and you have months of effort and no benefit. Start small and after two weeks you have something you would not give back.

Do not migrate your practice. Migrate one building, and let the rest run.

Step 1: one building, completely

Take a single building. Not the biggest, not the most complicated — one you know well and whose figures are clean. Twelve to thirty units is a good size: big enough to be real, small enough that a mistake does not cost you the day.

Bring that one building in completely: units, owners with their co-ownership shares (Miteigentumsanteile), opening balances, the last annual statement. Complete matters. A half-maintained building answers none of the questions you actually want answered — namely: does this hold up?

What you learn there, you know about your whole portfolio: how long capture takes per unit, where your old data has gaps, and whether the software works the way you work.

Step 2: the documents, not the accounting

The second step is the one nearly everyone skips, and it has the highest return per hour: the documents.

Digitising accounting is work with a late payoff — you only find out at the next statement whether it was worth it. Documents are the opposite. From the day the declaration of division, minutes, insurance policies and contracts of a building sit searchable in one place, a particular kind of phone call stops: "could you send me that again…".

It is also the step that reaches the owners. An owner portal, where every co-owner sees their building's documents themselves, removes exactly the work you could never bill for. That is the first point at which going digital is not merely tidier but cheaper.

Start with what is needed today — the last three years. The history behind that can follow later, or never; paper from 2011 is rarely asked for.

Step 3: the resolutions

Once a building is digital and the documents are in place, the Beschluss-Sammlung — the register of resolutions — is next. Not because it is fun, but because it is a statutory duty on which every handover turns. § 24 (7) WEG requires a consecutively numbered register; a manager who does not keep one finds out when an owner sues or a successor demands it.

The advantage of keeping it digitally is not convenience. It is that the numbering stops being your job. A register that arises from the resolutions themselves has no gaps and no numbers issued twice, because nobody has to issue them.

The detail — challenges, corrections, handover — has its own article: Keeping the register of resolutions right.

Step 4: only now the accounting

The Wirtschaftsplan (economic plan) and Jahresabrechnung (annual statement) come last, deliberately. They carry the hardest requirements, and they are where a mistake costs most — a cost type apportioned by the wrong key only surfaces when thirty owners are holding their statement.

Switch at the start of a financial year, never mid-year. A stub year across two systems produces a statement nobody can follow, yourself included.

And take opening balances for what they are: opening balances. Trying to re-enter five years of booking history is the single most common reason migrations peter out. The old statement stays valid even though it came out of another system.

Step 5: the rest of the portfolio

Only now the bulk — and now it is boring, which is the point. You know how long a building takes. You know which data you are missing. You have an order that worked.

Two recommendations:

• Sort by management contract, not by size. Buildings whose contract is up for renewal anyway are the natural next candidates.
• One building a week is fast enough. At forty buildings you are through in a year with nobody working late. Wanting to go faster should have a good reason behind it.

What this means for the old data

The point where it gets concrete is the export from your current software. Some vendors make it easy, some make it deliberately hard, and with some the export technically exists but is incomplete.

That is a topic of its own, with its own traps — we wrote it up here: Getting your data out of your old software.

One principle in advance: run every import as a dry run first. An import you cannot look at beforehand is not an import, it is a leap. And read the reconciliation before you commit, not after.

And the AI?

That comes last, and not out of scepticism. AI helps in property management exactly where structured data already exists: reading out invoices, suggesting document metadata, drafting replies. Pointed at a filing cabinet it can do nothing.

Which means steps 1 to 4 are the precondition for the AI part being worth anything at all. Start at the other end and you have bought a feature that cannot run on your data.

What actually works today and what is sales copy: What AI in property management can really do today.

The honest timeline

These are not marketing figures; they are the orders of magnitude practices actually move at. Anyone promising you a weekend has either not seen your portfolio or is not counting the data work.

What you do not have to do

Finally, the list of things sold as mandatory that are not:

• You do not have to digitise everything. Paper from ten years ago may stay paper.
• You do not have to treat every building the same. A three-unit community needs no portal.
• You do not have to re-enter the history. Opening balances are enough.
• You do not have to wait for a perfect system. The building that is digital today is worth more than the concept that will be finished next year.

The hardest step is the first, and it is smaller than it looks: one building, complete, in two hours. After that you are talking from experience instead of assumption.