The five things a guest judges
Everything a customer writes about a hotel fits into five headings. We did
not invent them: service quality research has been using them for forty years. Here they
are in plain language, each with a real sentence from your guests, and they change with
the filters above.
Where should you start?
The higher up, the more it moves your guests' rating. The further left,
the worse they rate it. Top left is what is urgent. The size of the circle is how
many people talk about it.
What they say, dimension by dimension
Share of mentions carrying a complaint. Every mention is backed by a
sentence the guest wrote: no quote, no count. Tap a row to filter the whole panel.
Positive
Mixed
Negative
Neutral
Where it happens
Share of mentions carrying a complaint for each dimension-and-department
pair. Below each one, how many guests mention it: without enough volume, the percentage
says nothing.
What it costs you in rating
The gap between the average rating of reviews where that dimension shows
up as a complaint and those where it shows up well rated. It is an association observed in
your own reviews, not a promise of what fixing it will add.
What if you improved it?
Move the sliders to where you believe you can get, and see where the index
would land. This is index arithmetic with today's measured weights, not a forecast: nobody
can promise you what will happen when you change something in your hotel.
What your best ratings hide
Four and five-star reviews that still contain a specific complaint. A
satisfied guest who points at a fault anyway: the cheapest criticism you will ever act on,
and the one that never shows up in the average.
How it moves over time
The reviews, as they are
The reviews matching the filters, with the sentences behind each dimension
highlighted inside the text itself. Green positive, red complaint, amber mixed. Guest
sentences are never translated: they are the evidence.
The three platforms, side by side
TripAdvisor, Booking and Expedia are not averaged together: they score on different
scales (1–5, 1–10 and 2–10) and different people write on them. The gap between them is
information, not noise to be smoothed away.
Next step
This is a photograph. The interesting part is the film.
What you have just seen is calculated on reviews already published. Connected to your channels, the same panel recalculates itself and warns you when a dimension starts to slip, weeks before it shows in the average rating, which is the slowest indicator there is.
RbQUAL does not tell you your rating has dropped: it tells you what is dropping it, in which department, and in which words from your guests.