Contents
Key facts
- Trade sources describe a well-run property in 2026 as four separate layers — the order tells you what to build first.
- Pricing has the clearest arithmetic: around USD 20 per property per month, usually covered within the first month.
- AI for guest messages pays off on the repetitive, not the urgent: a guest reporting a broken heater at 11 pm needs a fix.
- What we advise against: unedited AI listing texts and brightened photos — both raise an expectation the flat then disappoints.
The four layers
Trade sources describe a well-run property in 2026 as four separate layers, and that division is useful because it says in what order to build up.
First layer: a channel manager or management system that holds reservations and reconciles availability between the platforms. Second layer: a tool for pricing. Third layer: an AI layer for guest messaging. Fourth layer: an electronic lock for access without anyone present.
The order is not a recommendation but an observation: anyone starting at layer three is automating replies to enquiries about prices that are set wrongly. And anyone serving several platforms without calendar reconciliation risks double bookings — the most expensive mistake in this business, because a host cancellation damages the ranking for months.
Pricing: the item with the clearest calculation
PriceLabs is regarded as the market leader with more than 600,000 properties on the platform. The price is around USD 20 per property per month, and trade sources report that for most hosts the amount is covered by the additional income within the first month. In February 2026 a tool was added that reviews listing content and suggests improvements.
Wheelhouse relies more on competitive data and machine learning and is recommended for cities and event-driven markets — that is, precisely for Zurich with its trade-fair grounds, Kongresshaus and Hallenstadion. Its most striking feature is a slider between conservative, recommended and aggressive that controls how hard the algorithm pushes prices against the market.
Beyond is named for its simplicity. The choice can be summarised like this: PriceLabs for control, Wheelhouse for cities, Beyond for the simplest start.
What these tools cannot do: know a city’s events calendar in full. A congress affecting one industry drives prices in particular districts — an algorithm working from booking data recognises that only once others have already booked.
Your figures, not our examples
The calculator takes the number of rooms, the neighbourhood and the fit-out and shows you a range — with the arithmetic beside it, not as a single figure you have to take on trust.
Estimate the returnAI for guest messaging: when it pays
The names cited are Hospitable AI, Besty, Conduit and Hostella. The advice from the trade sources is restrained and therefore credible: add such a tool once the inbox becomes the bottleneck — not before.
The reason is a calculation. With one property and a few bookings a month, setting it up costs more time than it saves. From three or four properties the ratio tips: the number of similar questions — Wi-Fi password, arrival time, parking, the way from the station — grows faster than the number of properties.
What remains even with the best automation: the reply to a complaint. A guest writing at 11 p.m. that the heating does not work does not need a well-phrased message but a solution. That is exactly where the benefit of automation ends, and exactly where the review is decided.
What we advise against
AI-generated listing text without editing. It sounds recognisably like many others, and guests compare several listings side by side — where everyone writes the same thing, only the price decides. A text naming a peculiarity of the flat works harder than three paragraphs about a «central location».
AI-generated or brightened photographs. They raise expectations, and the disappointment on arrival shows up in the review. The connection is well documented: good photography raises the click rate considerably — but the effect rests on the flat matching the image.
Tools for replying to reviews. A reply to a poor review is read by future guests, and a visibly generated reply comes across worse than a short, honest one.
What we use
We name it because the question is legitimate: anyone paying for management wants to know how much of it is software.
For pricing we work with a tool, supplemented by the city of Zurich’s events calendar, which we enter manually — that is the part no algorithm supplies. Calendar management is centralised, not through mutual synchronisation between platforms: the detour via two channels querying each other is where gaps arise. Guest messages we answer with templates for recurring matters and personally as soon as something is at stake.
What we do not do: have replies to reviews generated, take listing text over unedited, retouch photographs.
Frequently asked questions
With continuous letting, yes — according to trade sources the additional income usually covers the roughly USD 20 a month within the first month. With a few weeks a year, no: there the setting up is not worth the effort.
Wheelhouse is recommended for cities and event-driven markets, which applies to Zurich with its trade-fair grounds and Hallenstadion. PriceLabs offers more control. Both work; what is decisive is that somebody knows the events calendar.
For recurring matters yes, for decisions no. A tool answers the question about the Wi-Fi password. It does not decide whether an enquiry is declined, whether damage is attributable to the guest, or what to do about a water leak on a Sunday.
And for your property?
Sources
State of our research: 24 June 2026. Official rules change — check the current position with the competent authority.
