A few days ago, while ordering coffee, the familiar screen appeared: “Add tip? 10%, 15%, 20%.” Without thinking too hard, I pressed “Yes.” I didn’t pause to consider the service or even the amount. I just didn’t want to click “No.” Maybe I didn’t want the barista or the person behind me in line to see me decline. Or maybe I simply followed the path of least resistance. Whatever the reason, that tiny digital prompt changed my decision.
As someone who loves thinking about human behaviour, I find this intriguing. Tipping is one of those areas where real life doesn’t quite match traditional economic theory. A rational consumer, acting purely out of self-interest, shouldn’t tip at all, especially when the service is already paid for. Yet people tip every day, and increasingly, we’re doing it digitally. Platforms like Uber Eats, Mr D Food, Checkers 60 delivery drivers, and point-of-sale systems in cafés now make tipping almost automatic by asking us directly before we complete payment. These small nudges are not random. They are designed with behavioural insights in mind, and they reveal a lot about what drives our decisions.
Research shows that the way tipping options are presented significantly changes how much people give. When digital payment systems include default amounts, such as 10%, 15% or 20%, those numbers act as anchors, suggestions that shape what seems “normal.” Evidence for this comes from a fascinating field study of over 13 million New York City taxi rides conducted by Kareem Haggag and Giovanni Paci (2014). When credit card machines in taxis began displaying default tip options, passengers overwhelmingly selected one of the preset amounts, and overall tipping increased. The researchers found that defaults acted as powerful behavioral cues: riders clustered around the suggested values, and average tips rose significantly compared to the old system where passengers had to manually enter a number. Interestingly, when the default percentages were set too high, a portion of riders opted to leave no tip at all, showing that while nudges can raise generosity, they can also backfire if perceived as unfair or excessive (Haggag & Paci, 2014).
In many cafés and delivery apps today, suggested tip percentages are now standard. As Harvard Business School researcher Jill Avery notes, digital payment systems have expanded tipping beyond traditional settings, and the presence of preset tip levels can influence what consumers perceive as a normal amount to give (Avery, cited in HBS Working Knowledge, 2023). These design choices may look small, but they subtly steer collective behaviour.
There is some suggestive evidence that tipping behaviour changes when the payment process feels public. For instance, a media report summarising U.S. data from 36,888 transactions claims that customers tipped more when the point of sale system was more visible (17.1% vs 15.8%) but that those using more private systems returned more often (StudyFinds, 2024). Research by Wang, Zhang and Wang (2025) found that when tipping requests appear in new contexts or before service is completed, they can provoke discomfort or resistance among consumers unless accompanied by visible service effort. The authors argue that this reaction reflects a tension between gratitude and perceived obligation, which could help explain why some digital tipping prompts feel awkward or intrusive.
When I tap “Add tip,” I sometimes imagine the driver or cashier noticing my choice. That imagined audience is powerful. Some might call it impression management, we act to protect our social image. Psychologists would say it’s about norm compliance, we don’t want to violate what we think everyone else is doing. Both explanations point to the same conclusion: the prompt works because it appeals not to rational calculation, but to social emotion.
In many ways, tipping prompts operate as a form of choice architecture. They do not remove freedom of choice, but by making certain options more prominent, easier, or more salient, they shift the path of least resistance. Thaler and Sunstein (2008) describe such interventions as “nudges”, small changes in how options are framed that influence decisions without forbidding alternatives. Digital tipping is often used as an example: for instance, the “Yes” or default tip option might be made more visually obvious or accessible, while declining or entering a custom tip may require more effort. In effect, these design choices push consumers toward the preset option.
The South African context adds another layer. Tipping is long-established in restaurants, hotels and taxis, where 10–20% is customary (Pawson, 2025). But with food delivery apps and cashless payments becoming more common, tipping norms are shifting online. There is not yet much local data on how digital prompts affect behaviour here, but it seems likely that the same nudges apply. It would be interesting to know how South African users respond to default options, or whether cultural attitudes toward fairness and guilt play a role.
While the economic framing helps explain part of the story, it is worth taking a lighter, more intuitive view. When we tip, we are not just transferring money. We are signalling something about fairness, gratitude and social connection. Still, some economic questions remain. Do these prompts lead people to pay more than they would like, creating a small welfare loss for consumers? Or do they serve a redistributive purpose, transferring income to low-wage workers who depend on tips? In that sense, tipping prompts might act like an informal form of redistribution, one driven by design rather than policy.
Delivery drivers and service staff often rely on tips as a significant part of their earnings, especially in low-wage markets. Many drivers are paid per delivery rather than hourly (JoziWire, 2022; University of Johannesburg, n.d.), meaning income fluctuates with demand, and tips can make up a crucial share of total earnings, particularly during peak times or in higher-income areas.
In South Africa, there are two quite different earnings logics at play and that matters for “who pays.” For restaurant and café workers, the National Minimum Wage Act applies: employers must pay at least the statutory minimum wage, and tips are legally “on top,” not a substitute for base pay. Put simply: in restaurants, the employer is on the hook for the wage floor; tips are an add-on from customers, not a replacement (Department of Employment and Labour, 2023).
Platform delivery is different. Couriers are typically paid per task (not an hourly wage), so earnings swing with order volume, time of day, distance and location (MotionAds, n.d). Platforms also emphasise that tips are consumer-funded and passed through: Uber states that 100% of tips in South Africa go directly to the driver/courier (Uber, n.d.). Driver-facing guidance and local reporting similarly describe per-delivery pay with boosts and tips varying by area and peak times, reinforcing that income is demand-driven rather than guaranteed hourly pay (JoziWire, 2022). Fairwork’s South Africa ratings show why this distinction matters: after fuel, data and other costs, some platform workers’ take-home earnings can fall below the minimum wage benchmark, which means tips can function as a material share of take-home pay rather than a small extra (Fairwork, 2021).
There is also the risk of what some have called “tipflation.” (Tipflation is the increased percentages of tips suggested by new technology). As digital systems normalise tipping in more places, people start to feel fatigued or uncertain. A 2023 Pew Research Center survey found that 72% of Americans believe tipping is expected in more places than before, and many are unsure when and how much to tip. The default amounts have quietly increased, shifting the social norm upward and leaving consumers feeling pressured (Pew Research Center, 2023).
Despite the downsides, there are reasons to see tipping prompts as an example of behavioural design working toward good outcomes. If digital nudges raise average income slightly without creating resentment, that could be seen as welfare-enhancing. But transparency matters. People tend to accept nudges when they feel fair and resist them when they feel manipulative.
I often wonder what might happen if we experimented with different designs in South Africa. What if delivery apps randomly varied the default tip options to see how behaviour changes? Or if cafés displayed average tip data anonymously to make norms explicit? What if apps showed how tips are distributed between drivers and the company? These are small tweaks, but they could reveal whether we’re tipping because we genuinely want to reward service, or simply because we feel nudged into it.
Every time I see that question, “Add tip?” I usually don’t pause right then. I tap, move on, and only think about it later. It’s often afterward, when I’ve already paid, that I catch myself wondering why I tipped, was it gratitude, guilt, or simply habit? Behavioural economics doesn’t judge these motives, it simply shows how they coexist. And perhaps that’s the broader lesson: in a world where apps increasingly shape our decisions, the smallest design choices can have large effects on fairness, dignity and everyday generosity.
References
Avery, J. (2023). Technology and COVID upended tipping norms: Will consumers keep paying? Harvard Business School Working Knowledge. https://www.library.hbs.edu/working-knowledge/covid-and-technology-upended-tipping-norms-will-consumers-keep-paying
Haggag, K., & Paci, G. (2014). Default Tips. American Economic Journal: Applied Economics, 6(3), 1–19. https://www.kareemhaggag.com/f/Default_Tips.pdf
JoziWire. (2022). How much does an Uber Eats driver or rider earn in South Africa? JoziWire.co.za. https://joziwire.co.za/2022/07/how-much-does-an-uber-eats-driver-or-rider-earn-in-south-africa/
MotionAds. (n.d.). How much do food delivery drivers really earn in South Africa? MotionAds.
Pawson, N. (2025). Tipping in South Africa – When and how much. https://www.wisemove.co.za/post/tipping-in-south-africa-when-and-how-much
Pew Research Center. (2023). How Americans see recent developments in tipping. https://www.pewresearch.org/2023/11/09/how-americans-see-recent-developments-in-tipping/?gad_source=1&gad_campaignid=22378837192&gbraid=0AAAAA-ddO9Fu7pFpT9zGa-nFzS76D7k53&gclid=CjwKCAjwr8LHBhBKEiwAy47uUo3VidvAokdTClx0ftX5GsC1BmaWrgcyn7x7f5_H0xoBHWwUoDXsxBoCTdAQAvD_BwE
StudyFinds. (2024). ‘Guilt tipping’: Visible tip screens make customers pay more — but they might not come back. StudyFinds.org. https://studyfinds.org/guilt-tipping-changing-consumers/
Thaler, R. H., & Sunstein, C. R. (2008). Nudge: Improving decisions about health, wealth, and happiness. Yale University Press. https://www.researchgate.net/publication/257178709_Nudge_Improving_Decisions_About_Health_Wealth_and_Happiness_RH_Thaler_CR_Sunstein_Yale_University_Press_New_Haven_2008_293_pp
Uber. (n.d.). Adding extra amount for Delivery Partners (Uber Eats tips in South Africa). https://about.ubereats.com/za/en/how-it-works/delivery-tipping/
University of Johannesburg. (n.d.). The gig economy and food-delivery in South Africa: Opportunities and challenges. UJContent. https://ujcontent.uj.ac.za/esploro/outputs/graduate/The-gig-economy-and-food-delivery-in/9942805207691
Wang, X., Zhang, Y., & Wang, H. Rethinking tipping request: Examining consumer reactions in emerging tipping contexts. (Paper). Retrieved from https://www.researchgate.net/publication/391453099_Rethinking_tipping_request_Examining_consumer_reactions_in_emerging_tipping_contexts