Server or cloud: we don't know whether to move our PMS

The server sits in the office or at the IT contractor. Who installs updates, checks backups and reacts to an outage is not defined. As long as nothing happens, nobody notices. In the ÖHV technology survey 2025, 42% of surveyed properties still ran server-based, many of them in exactly this situation.

Write it down honestly: who maintains, who backs up, who is reachable at night, what it costs per year. If the answer to two of those is “nobody”, the property is already a cloud candidate, regardless of vendor.

The worry: no internet, no system; the data sits “somewhere”; the vendor can raise prices at will. Behind it is often a bad experience with some other online service, not an analysis of the own operation.

Check each worry on its own: how often did the internet fail last year, and for how long? Where are the vendor's data centres, and what does the contract say about data export and price lock? Every answer in writing. Gut feeling turns into a list you can negotiate with.

One module runs in the cloud, the PMS on the server, the POS somewhere else again. Every interface is its own point of failure. 25% of ÖHV properties ran hybrid in 2025, mostly not by plan but as the result of one-off decisions.

Define a target architecture, even if getting there takes three years: which systems should end up where? Every new purchase is checked against that target, not against the current offer.

“If we go cloud, we'll switch the whole PMS at the same time.” Two projects become one, and the property pushes both ahead of itself because it looks too big.

Separate them: if the current vendor offers a cloud version, the move is a manageable project with known data. A system switch is its own decision with its own requirements list. First decide whether the PMS stays, then where it runs.