Measuring customer satisfaction: kiosk and web, not one or the other
25 juillet 2026
Two channels, two biases
A survey emailed three days after a visit doesn't measure the visit. It measures the memory of the visit, filtered through the customers who bother to reply, usually the very happy and the very unhappy.
A kiosk at the exit measures something else: the immediate reaction, including from lukewarm customers who would never have opened an email. In exchange, it captures poorly what happens afterwards (delivery, after-sales, invoicing).
Neither channel is "the right one". Pitting them against each other just means picking your bias.
What that implies for the tool
If the two channels are complementary, the tool has to treat them as one survey rather than two separate products. Concretely:
- The same survey runs on the kiosk and via link or QR code. No duplicated questionnaire, no risk of the two versions drifting apart.
- Results land in the same dashboard, with the channel as an analysis dimension rather than a silo.
- Question structure is identical, which keeps cross-channel comparisons readable.
That's the position SatisfactionZ takes: one platform, two collection points.
What makes a kiosk survey work
Response rate on a kiosk depends less on the survey content than on the physical context.
Three questions maximum. Beyond that, drop-off climbs sharply. A rating, a reason, an optional free-text field: that's a format that holds.
A first screen you understand without reading. Smileys or a visual scale land better than a sentence, especially with a multilingual clientele.
A position on the natural exit path. A kiosk tucked in a corner collects nothing, however good the application is.
No identity to enter. Asking for an email drops participation and adds nothing if the goal is to track a trend per site.
The under-discussed part: multi-site
As soon as there's more than one outlet, the problem changes nature. It stops being about collecting responses and becomes about who sees what.
The questions that come up in practice:
- Should a site manager see the other sites' results?
- Who can edit a survey that's already live, and what happens to responses already collected?
- How do you compare two sites with completely different footfall?
These get settled during scoping. A tool that only handles roles at the organisation level becomes unusable past three or four sites.
Measuring isn't enough
A satisfaction dashboard that only gets looked at in a quarterly meeting changes nothing. What produces an effect is short loops: a low score repeating on a specific time slot, raised with the team concerned within the week.
That's why real time isn't a marketing argument here: it's the condition for the data to be worth anything.
The demo is freely available on the SatisfactionZ page. To discuss it, contact us.