Cookie settings

Choose what we may store on your device. You can change your choice at any time with the link at the bottom of every page.

Questions about your data? Contact us

Running a restaurant

Kiosks, QR and online ordering: what the evidence says

What published research says about self-ordering, order size and speed, where a kiosk, a QR menu or your own webshop helps, and what to check before you decide.

By Resto team Published Updated 6 min read

In this guide
  1. 01What the studies found
  2. 02What this means for a restaurant
  3. 03Where self-ordering often fits
  4. 04Where it fits less well
  5. 05Your own channel or a delivery app
  6. 06Allergens before payment
  7. 07Before you decide

Kiosk and app sellers often say that guests spend around 30 % more when they order on a screen. We could not find a published study with a clear method behind that number. A few studies do exist. This guide explains what they measured, what they found, and how you can test the question in your own restaurant. The research is as of 2 October 2026.

What the studies found

Online pizza orders, peer-reviewed

Four researchers studied one US pizza delivery restaurant that added online ordering next to phone and counter orders. The data had 160,168 orders from 56,283 customers, and 6.7 % of the orders came in online.1 The study was published in Management Science in 2015.1

When the researchers compared the same customers before and after online ordering, the average item in an online order had 14 % more special instructions, such as half one topping and half another, and 3 % more calories.1 In the simple averages, web customers spent $0.61 more per order, because they added more toppings.1

The authors explain the result with social pressure. One of them, Ryan McDevitt, said: “Online, you’re not making anyone wait while you place your special order. No one’s judging you.”2

Kiosks in 20 fast-food restaurants, university study

Researchers at the University of Hamburg worked with a fast-food company that had kiosks next to its counters. They used transaction data from 20 of its restaurants and surveyed guests in 2 of them.3 In their two studies, guests spent 14 % and 16 % more at the kiosk than at the counter.3 The kiosk orders also had 32 % more sugar and 42 % more trans fats.3

The researchers give two causes. The kiosk always offers extras, such as a coffee with the meal. And guests feel less observed when they order unhealthy food.3 We found this study only as an interview on the university website, so we do not know the full method. Also, guests chose the kiosk or the counter themselves, and the guests who choose a kiosk can be different from the guests who choose the counter.

One Burger King in a student union, master’s thesis

A master’s student at the University of North Texas randomly sent 192 guests at a Burger King to the kiosk or to the cashier.4 Ordering at the kiosk took 48.9 seconds on average. Ordering at the cashier took 58.7 seconds. The kiosk was about 17 % faster.4 But kiosk guests spent $5.56 per person, against $6.26 at the cashier, which is about 11 % less.4

The author suggests three possible reasons: students who want to save money, the design of the kiosk screen, and cashiers who sell extras.4 This is a thesis and not a peer-reviewed paper. The sample is small, it comes from one restaurant, and the guests reported their own spend in a survey.

What this means for a restaurant

The studies agree on one thing. On a screen, guests take their time, change dishes more, and choose things that they might not say out loud.

The studies do not agree on order size. Where the kiosk offered extras every time, orders grew. In a student setting where cashiers sold extras, kiosk orders were smaller. So the design of the screen and the habits of your staff matter more than the device itself. No study gives a number that you can safely put in your budget.

University of Hamburg, first study+14 %University of Hamburg, second study+16 %Master’s thesis, one Burger King (192 guests)−11 %
The studies used different methods. In Hamburg, guests chose the kiosk or the counter themselves. In the thesis, the researcher sent guests to one or the other at random, and guests reported their own spend. The pizza study gives dollars per order, not a percentage, so the chart does not show it. Source: Universität Hamburg3 and the University of North Texas.4
Show the data
StudyDifference in spend
University of Hamburg, first study+14 %
University of Hamburg, second study+16 %
Master’s thesis, one Burger King (192 guests)−11 %

Of these studies, only the thesis measured speed, and it found the kiosk faster. That result comes from one restaurant, so treat it as a good sign rather than a fact.

Where self-ordering often fits

  • Counter service with queues at peak times. Fast food places and food halls can take several orders at the same time when there are several kiosks.
  • Many guests who do not speak Icelandic. A guest can read the menu in their own language and take the time they need.
  • Menus with many choices. The pizza study shows that guests change dishes more on a screen. The kitchen then gets the instruction exactly as the guest wrote it.
  • The second round at the table. With QR ordering, a table can order another drink or a dessert without waiting for a server.

Where it fits less well

  • Table service where advice is part of the meal. A server who suggests a wine or explains the fish of the day gives something that a screen cannot.
  • Guests who prefer to talk to a person. Keep a counter or a staff member available, so that each guest can choose.
  • A screen with a weak design. The thesis names the kiosk design as a possible reason for its smaller orders.

Your own channel or a delivery app

A delivery app brings guests who do not know you yet, and it brings couriers. Wolt, for example, takes a commission on each order that comes through its platform. Signing up is free, and you can end the agreement at any time.5 Wolt does not publish its commission rate on that page, so ask for the rate and all other fees in writing.

Your own webshop or QR menu keeps the full relationship with the guest. You set the menu and the prices, and you see who orders. You also handle pickup or delivery yourself. Many restaurants use both: the app for new guests and their own channel for regular guests.

Allergens before payment

The allergen rules apply to every channel. For online orders the allergen information must be available before the purchase is completed. On a kiosk or a QR menu, the guest makes the choice on a screen, so the allergens belong on the screen too. Read more in our guide Allergens on the menu: what Icelandic rules require.

Before you decide

  1. Measure today. Take the average order value and the number of orders per hour at peak times from your POS reports, for at least four weeks.
  2. Test on a small scale. Try one kiosk, or QR ordering on one floor, and compare the same weeks and days.
  3. Look at what the screen offers. Check which extras it suggests and when, and decide which ones you want.
  4. Ask for the method. When a seller quotes a number, ask how many restaurants, which period and what it was compared with.
  5. Check the allergens and the languages. Make sure that each dish shows its allergens before the guest pays.
  6. Get the delivery app costs in writing. Calculate what you keep from an average order after commission and fees.

If you use Resto, the kiosk, QR ordering and online ordering use the same menu as the POS. You can try one channel in one place and compare the reports before and after.

Sources

  1. The Effect of Social Interaction on Economic Transactions: An Embarrassment of Niches? (working paper, December 2012; published in Management Science 61(12), 2015) Goldfarb, McDevitt, Samila and Silverman · Accessed
  2. How Self-Service Influences Customer Behavior (3 August 2015) Duke University, Fuqua School of Business · Accessed
  3. Effects of self-service kiosks on buying behavior (interview with Prof. Karen Gedenk, 2023) Universität Hamburg · Accessed
  4. Examining Self-Service Kiosks in Quick-Service Restaurant Settings (master's thesis, 2019) Bryan C. Torres, University of North Texas · Accessed
  5. Wolt for restaurants Wolt · Accessed

More guides

Book a demo of Resto

We show you the system at a time that suits you. The demo does not commit you to anything.

We read every request and get back to you ourselves. We also answer in the evening and at the weekend.

hello@resto.is

We use these details only to contact you about Resto.