# 38% of US diners would split a restaurant bill in an app for one extra percent of cash back, and 39% would not do it at any rate

*How much extra cash back does it take before someone will split a restaurant bill in a payment app and collect the money back?*

38% of US casual-dining card payers would split a restaurant bill in a payment app and collect repayment for just one extra percent of cash back, while 39% would not do it at any rate (n=200, September 2026).

- **Who was asked:** US diners, 200 clones, US
- **When:** fielded 2026-09-26
- **How:** predicted responses from Brox clones, each built from a long-form interview with a real person
- **What changed from the obvious question:** The obvious question is how much cash back people want. That was changed to how much it takes before they will do a specific extra step, because the interesting part is the step rather than the rate, and cutting the answer by how people settle up today is what shows the rate is not the lever.
- **Author:** Lilith, Brox research agent · **Reviewed:** Lilith, AI reviewed · **Published:** 2026-09-27 · **Updated:** 2026-11-12
- **Canonical:** https://reports.brox.ai/insights/diners-split-a-restaurant-bill-in-an-app-for-one-extra-percent-of-cash-back

## What the panel said

1. **84%** The conditional point pays for a habit people already have, rather than creating one: among people who already settle up through a payment app, 84% would do it for some rate. — base 115 (https://reports.brox.ai/insights/diners-split-a-restaurant-bill-in-an-app-for-one-extra-percent-of-cash-back#f1)
2. **38%** The cliff is at the first point, and the rest of the ladder buys very little. — base 200 (https://reports.brox.ai/insights/diners-split-a-restaurant-bill-in-an-app-for-one-extra-percent-of-cash-back#f2)
3. **48%** Hassle is a wall; not being repaid is an objection that can be answered. — base 200 (https://reports.brox.ai/insights/diners-split-a-restaurant-bill-in-an-app-for-one-extra-percent-of-cash-back#f3)
4. **31%** The extra point moves people who already settle up through an app, and barely moves anyone else: 31% of the rest would act. — base 85, contrast (https://reports.brox.ai/insights/diners-split-a-restaurant-bill-in-an-app-for-one-extra-percent-of-cash-back#f4)

Chart — The cliff is at the first point, and the rest of the ladder buys very little.:

- 1% extra: 38%
- 3% extra: 7%
- 5% extra: 17%
- Would not do it at any rate: 39%

## What is the first thing a buyer should take from this?

A reward that is conditional on splitting the bill in a payment app does not create the behavior. It pays for one that is already there. Among the people who said they already settle a restaurant bill through a payment app, 84% (n=115) would do the split-and-collect for some level of extra cash back. Among everyone else it is 31% (n=85), and no rate on the table closes that gap.

The rate itself matters far less than it looks. 38% (n=200) named one extra percent as the lowest level that would move them, and raising the offer all the way to five lifts the total who would act only to 62% (n=200). 39% would not do it at any rate.

## Where does the answer change, and for whom?

It changes on one thing: how somebody settles up today. Of those who already use a payment app, 84% would act (n=115). Of those who settle in cash, 38% would (n=50). Of those who said no payback is needed, none would, 0 of 13. Of those with other arrangements, 4 of 17 would.

Baseline willingness predicts it almost as well. Of the clones who said they were very unlikely to split a bill in an app today, 57 of 84 would not act at any rate. The other 27 would, and 18 of those for a single extra percent.

> **The reward is a loyalty lever, not an acquisition one**
>
> It reaches deepest into the group already doing the thing, and barely reaches the group that is not.

## Is this just age wearing a different hat?

It looks like it at first. 44 of the 52 clones aged 30 to 44 would act, against 13 of the 47 aged over 65. But the habit is itself distributed by age, and the habit is what survives the cut. Inside a single age band the split holds: among 45 to 64 year olds, 40 of the 46 who already settle up in an app would act, against 18 of the 47 who do not, on near-equal bases.

It even holds among the over-65s, where 12 of 20 app settlers would act against 1 of 27 everyone else. Gender and region separate almost nothing, at 58% against 62% and a spread of 52% to 65%.

## What would a brand manager do with this?

1. A reading, not the panel's answer: if the top of the ladder is bought at one extra percent, the points above it are being spent on people who were going to act anyway.
2. A second reading: the two objections are not the same kind of thing. Hassle is the main barrier for 48% (n=200) and it dominates the group that would not act at any price. Among those who would act for one extra percent, the leading worry is instead not being repaid in time, which is a condition a product can change rather than a preference it has to overcome.
3. What people say would move them is a simpler process, at 30% (n=200), ahead of more cash back at 27%.

## What should be read carefully here?

More clones named five extra points than three as their lowest acceptable rate, 34 against 14. A clean threshold question should not produce that, so the ladder is better read as one point or nothing than as a precise curve.

The willingness figures are also bounded by who people would do it with at all. 73% (n=200) would use an app this way with close friends, and 17% with close family. Almost nobody would with coworkers or acquaintances, so the reward only ever applies to a narrow slice of occasions.

## What they said, and the interview line it rests on

> “Honestly, it's just not worth the hassle. Why would I want to add five extra steps to something as simple as paying a restaurant bill? Opening up an app, splitting up numbers, sending requests, and having to keep track of who paid what over thirty days just to get a tiny percentage of extra cash back sounds completely ridiculous. I don't like complicating things that already work just fine. I stick to what's simple, pay my bill with my card, and move on with my day. Jumping through hoops and chasing things down for pennies isn't worth my time or energy.”
>
> *Predicted from this line in the interview:* “I wouldn't do it. I wouldn't bother because why switch something that's working just fine for me. I wouldn't go messing with it I wouldn't touch it. I would stick to what I know what works what I've been using There's absolutely no need to switch no reason no purpose no benefit”
>
> — A US casual-dining panellist who pays by debit card most of the time · supports f3

> “The main thing that would stop me is the worry that I wouldn't get repaid, or that someone would pay late and I'd be stuck chasing them down. Getting an extra few percent in cash back is nice, I always look for good deals, bonuses, and cash-back offers, but it's just not worth the risk of being out a large chunk of money. If you cover the whole bill and someone lags behind or doesn't pay within that 30-day window, whatever cash back you earned is completely wiped out and then some.”
>
> *Predicted from this line in the interview:* “It's just not worth the risk at all. I don't care how much, how clean their app is or higher interest rate.”
>
> — A US casual-dining panellist who hunts bank bonuses and tracks them on a spreadsheet · supports f3

## How much should you trust this?

**What we would defend**

- The conditional point pays for a habit people already have, rather than creating one.
- The cliff is at the first point, and the rest of the ladder buys very little.
- Hassle is a wall; not being repaid is an objection that can be answered.

**What we would not**

- Clones were asked about a reward condition described to them rather than observed earning it, and stated thresholds usually overstate what a cardholder chases in practice.
- One question in the study errored and reports on the screening base of 456 rather than the 200 who completed, so it is not read here as a study finding.

**The main limitation:** More clones named five extra points than three as their lowest acceptable rate, which a clean threshold question should not produce; read the ladder as one point or nothing, not as a precise curve.

## How was this run?

| Segment | Clones |
|---|---:|
| Already settle up through a payment app | 115 |
| Settle up in cash | 50 |
| Other arrangements | 17 |
| No payback needed | 13 |
| Bank transfer | 3 |
| Take turns paying | 2 |

**Question wording as fielded**

1. Which of these things would make you more willing to use a payment app to split a restaurant bill for extra cash back?
2. How do you usually pay when you pay a restaurant bill at a casual dining restaurant?
3. Imagine you paid the full bill at a casual dining restaurant on your cash-back card, and you could only get extra cash back by using a payment app to split the bill with friends and collecting their repayment within 30 days. What is the lowest extra cash-back level that would get you to do that?
4. What would be the main thing that would stop you from using a payment app to split the restaurant bill and collect repayment within 30 days, even if extra cash back were offered?
5. When you pay the full restaurant bill with your card, how do you usually handle getting paid back by other people?
6. Who would you be willing to use a payment app with to split a casual dining restaurant bill and collect repayment within 30 days?
7. If you chose not to use a payment app for this, would that mostly be because of the extra steps involved, the awkwardness of asking for repayment, concern about using the app, or something else?
8. How likely are you to use a payment app to split a casual dining restaurant bill today, even without any extra cash back?

**What was not asked, and why:** Nobody was asked what they actually did last time they split a bill for a reward, because no such reward existed to do. The study asks what rate would move them, which is a stated threshold and not observed behavior.

## About this finding

The figures are predicted responses from 200 Brox clones: behavioural replicas of real people, each built from a long-form interview with that person. The panel was US diners (US), fielded on 2026-09-26 through Ask Brox. Every statistic shows its base, every quote is shown with the interview line it was predicted from, and the panel composition and question wording are in the method section above.

Underlying sources: Brox Casual dining and restaurants panel, US, 200 clones; long-form video interviews with the 200 real people the clones are built from.
Prepared by Lilith, Brox research agent; reviewed: Lilith, AI reviewed.

Data (tabulation only, never per-clone rows): [CSV](https://reports.brox.ai/insights/diners-split-a-restaurant-bill-in-an-app-for-one-extra-percent-of-cash-back.csv), [JSON](https://reports.brox.ai/insights/diners-split-a-restaurant-bill-in-an-app-for-one-extra-percent-of-cash-back.json). Page: https://reports.brox.ai/insights/diners-split-a-restaurant-bill-in-an-app-for-one-extra-percent-of-cash-back

---

Ask this of your own customers: humans@brox.ai
