Show short descriptions on kiosk category screens when guests cannot reliably identify dishes from their names alone. Show more items at once when names and pictures already make the choices clear. The right decision is the one that helps first-time guests find and recognize the right item without making the category hard to scan.
Do not decide from a screenshot alone. Ask a vendor to demonstrate both layouts with your menu, then watch first-time users complete the same browsing tasks. Measure wrong selections and requests for help as well as time. A screen that looks faster to an experienced manager may leave a new guest guessing.
Start with a category guests may not know
Pick a real category with several unfamiliar names, similar-looking dishes, or ingredients that matter to the choice. A category of plainly named drinks may not tell you much. A group of house bowls with names such as Sunrise, Market, and Ember is more useful: those names do not tell a new guest which bowl contains chicken or which is meat-free.
Bring the category's item names, approved descriptions, and current prices from your menu. Ask the person who maintains the menu to check that the descriptions match what the kitchen actually serves. For this decision, concentrate on browsing and recognition, not the later design of modifier screens.
Separate what you know from what you are assuming. You may know how many items are in the category and what the menu says about each one. You do not yet know whether a first-time user will understand those names on a kiosk, notice a short description, or find the next screen. Those questions need a demonstration.
Hypothetical worksheet assumptions: A restaurant has eight house bowls. Their names are unfamiliar to new guests, and the approved menu identifies the main ingredients. Layout A shows four items at once with descriptions of up to eight words. Layout B shows all eight at once with names and pictures but no category-screen descriptions. Both allow a guest to open an item for more detail. The restaurant wants users to find the intended bowl quickly and select it correctly.
These are comparison inputs, not claims about a particular kiosk. Ask each vendor whether it can configure both views, how much text fits on the proposed hardware, and what happens when a description is longer than the available space.
Compare density with understanding
More visible items can reduce browsing when guests already recognize what they want. It can also make the choices look crowded or force text and pictures to become too small. Descriptions take space, but a few useful words may prevent guests from opening several items or choosing the wrong one.
Use the worksheet below to compare two possible results for the same hypothetical category. The figures are illustrative outcomes, not results from a completed test. Each outcome assumes 12 first-time users try each layout on comparable tasks. A correct selection means the user opens the intended item; the time runs from the category screen to that selection.
| Measure | Layout A: descriptions | Layout B: more items |
|---|---|---|
| Items visible at once | 4, with up to 8 description words each | 8, without category descriptions |
| Outcome 1: correct selections | 11 of 12 | 8 of 12 |
| Outcome 1: median time | 24 seconds | 19 seconds |
| Outcome 2: correct selections | 11 of 12 | 11 of 12 |
| Outcome 2: median time | 25 seconds | 18 seconds |
In outcome 1, choose A for this category if the restaurant values correct identification over the five-second median difference. B is faster, but four of 12 users select the wrong item. That could create questions or corrections later in the order. Do not assume a wrong selection reaches the kitchen; record where the user catches it.
In outcome 2, B is the better candidate: both layouts have the same number of correct selections, while B has a shorter median time. Check that its text and touch targets are still usable before accepting it. The result changes because added descriptions no longer appear to improve selection accuracy in this example.
Write descriptions for the decision a guest is making. “Chicken, rice, avocado, lime dressing” is more useful for identifying a bowl than a string of promotional adjectives. Keep important distinctions near the beginning, and avoid treating a short category description as a complete ingredient or allergen statement. Make fuller item information available where guests can find it, and give them a way to ask staff when they need help.
Run a safe demonstration and catch the failure case
Give the vendor the same menu content and task list for each layout. Ask it to use the proposed screen size and show both views at the text size a guest would actually see. If the vendor cannot reproduce a layout, record that as a limitation rather than comparing it with a mockup as if both were working screens.
Recruit people who do not already know the category. A useful task is: “Find the bowl with chicken and lime dressing.” Do not point to the right answer or explain the category names. Start timing when the category appears; stop when the participant opens the item they believe matches. Record the selection, time, requests for help, and any visible difficulty reading or reaching a control.
Vary task order and which layout each person sees first. Otherwise, someone may recognize the answer in the second layout simply because they learned it in the first. Use different but comparable item-finding tasks, and record which task and layout each person receives. With a small group, treat the result as a buying signal, not a forecast of every guest's behavior.
The shift manager should observe without coaching. If a participant asks for help, the manager records the request and gives the same neutral assistance defined before the demonstration, rather than naming the item. This prevents a helpful staff member from making a confusing layout look easier than it is. A menu owner can review any wrong selections afterward to distinguish unclear wording from an incorrect task answer.
Keep this a simulation. Stop at item selection, or use a vendor-provided practice environment for any later steps. Do not require a real card payment, send a guest message, or dispatch a courier. If you later run a live service pilot, plan its staffing, actual order handling, payment setup, and fallback separately; it is not the same as this practice check.
Set a pass or fail rule before the demo
For the hypothetical eight-bowl category, the owner might set this rule: at least 11 of 12 participants must open the intended item, and the median selection time must be no more than 30 seconds. Both layouts must also show readable text and usable controls to the participants. Under that rule, Layout B fails outcome 1 on accuracy, while both layouts pass the numerical thresholds in outcome 2. The owner can then favor B's greater density if the readability and control checks also pass.
Choose thresholds that fit your own service. Write them down before seeing results, along with what counts as an error and how you will handle a participant who needs help. A median hides individual struggles, so review the slowest attempts and help requests too. A passing median should not erase a recurring problem for guests who cannot read or use the screen.
Use this checklist when you ask for the demonstration:
- Menu content: Can the vendor enter the real category names and descriptions, and show how long text appears?
- Layout: How many items fit at a usable size with and without descriptions? Can different categories use different approaches?
- Detail: How does a guest open fuller item information and return to browsing? What happens if a description changes?
- Order flow: After selection, can staff and guests review the choice before an order is sent? Ask to see the proposed setup rather than assuming every workflow is available.
- Fallback: How can a guest get staff help, and how will staff enter or correct an order if the kiosk cannot be used?
Assign the menu owner to resolve wording errors and the shift manager to own service exceptions. If a layout fails, change the description or density, then repeat the same kind of simulation with new first-time participants. Keep the original result and the revised result separate.
Before signing, compare written quotes for hardware, software, installation, and payment compatibility alongside the screen result. Ask what is fixed, recurring, or charged only for exceptions such as an additional visit. A quoted processing fee alone is not the all-in operating cost. Check Fleksa's current plans for its published pricing; pricing information was checked in November 2026. For a broader cost framework, see the kiosk break-even guide.
Also ask how kiosk orders fit your restaurant POS system and kitchen workflow. Fleksa's kiosk cart management guide is a useful starting point for order review questions. Confirm the exact setup in the demonstration.
Next step: Take one difficult category, its approved descriptions, and your pass rule to a Fleksa kiosk demonstration. Compare your requirements with the available setup and confirm current support before choosing a workflow.
Frequently asked questions
Should every kiosk item have a description on the category screen?
No. Use category-screen descriptions where they help guests distinguish unfamiliar or similar items. Clear names and pictures may be enough elsewhere, provided guests can find fuller information when needed.
How many items should a kiosk screen show?
There is no universal count. Compare layouts on the actual screen size, checking whether guests can read the content, use the controls, and select the intended item without unnecessary help.
Is a faster layout always the better choice?
No. Compare time with correct selections and help requests. A few seconds saved may not be worthwhile if guests repeatedly choose the wrong dish or need staff to explain the menu.
