Contents
- 30-second summary: the seven key takeaways
- The trial that divided the couple of elements
- Notifications do not affect everyone in the same way
- How much nudges amplify risk
- The study with two portfolios at once
- Which part of behaviour it acts on
- Practical steps to apply this
- Frequently asked questions
- What we do not know
- References
The confetti when a trade completes, the achievement badges and the trending notifications have become the favourite target of regulators and headlines. The most rigorous randomised experiment does find an effect: hedonic gamification increases trading volume by 5.17 %. But it also finds something that is almost never quoted: 70 % of the difference between platforms is not caused by gamification, but by who chooses each platform. And there is a study with real and simulated portfolios at the same time in which playing reduced overconfidence and the disposition effect while increasing two other biases. This article sets out what has actually been measured.
30-second summary: the seven key takeaways
Each row names the study and the year. It is worth separating the effect of gamification from the effect of who it attracts.
| Finding | Result | Study and year |
|---|---|---|
| Impact of hedonic gamification on trading volume | Averaging +5.17 % | Chapkovski et al., Management Science (2024) |
| Origins of the platform disparity | Driven by 70 % self-selection alongside 30 % gamification | Chapkovski et al. (2024) |
| Demographics favoring platforms featuring confetti | Users exhibiting lower financial literacy | Chapkovski et al. (2024) |
| Impact of behavioral nudges concerning risk | They tend to amplify exposure, particularly during high volatility | Chapkovski et al., J. Behav. Exp. Finance (2025) |
| Degree of protection provided by financial literacy | A single standard deviation reduces the impact by 56 % | Chapkovski et al. (2025) |
| Biases mitigated by a stock market game | Overconfidence together with the disposition effect | Şenol and Onay, J. Behav. Exp. Finance (2022) |
| Biases exacerbated by it | Familiarity alongside the status quo | Şenol and Onay (2022) |
What you will find here
• The randomised experiment that separated the effect of the design from the effect of who chooses it.
- What trending notifications do exactly, depending on the beliefs the user arrives with.
• How much nudges amplify risk-taking, and what holds it back.
- The study involving a live portfolio alongside a game portfolio simultaneously, and its surprising outcome.
• Which part of financial behaviour gamification actually acts on.
The trial that divided the couple of elements
Philipp Chapkovski, Mariana Khapko and Marius Zoican released the benchmark study in Management Science: a randomized online experiment exploring how gamification influences retail investor behavior.
Their initial discovery concerns decision-making dynamics: individuals possessing lower financial literacy gravitate toward platforms featuring hedonic gamification components, including achievement badges and confetti.
The second is the average effect: on average, hedonic gamification increases trading volume by 5.17 %.
And the third is the one that changes the interpretation of everything else: the difference in activity between gamified and non-gamified platforms is mainly due to self-selection, at 70 %, rather than to gamification, which accounts for the remaining 30 %.
Why that distinction matters so much
Comparing the activity of users of a confetti app with that of a sober one does not measure what it appears to. It mostly measures the difference between two kinds of people who were already different before installing anything. The study itself documents this in behaviour: participants who prefer hedonic gamification display noisy trading strategies, while those who prefer non-gamified platforms show more contrarian behaviour.
| Study | What it analysed | What it found | What it does not support |
|---|---|---|---|
| Chapkovski et al. (2024) | Randomised experiment featuring gamified platforms | A 5.17% volume increase; self-selection accounts for 70% of the impact | An experimental setup rather than actual funds at a brokerage |
| Chapkovski et al. (2025) | 605 participants across four nations, utilizing a virtual asset | Prompts heighten the level of risk assumed | A simulated asset; actual financial losses remain unmeasured |
| Şenol and Onay (2022) | Investors managing both real and simulated portfolios simultaneously | Two cognitive biases decline while another pair increases | Voluntary engagement in the game: self-selection persists |
| Agrawal and Sharma (2025) | 557 financial application users situated in India | Planning and self-efficacy are moderated, whereas attitude is untouched | Cross-sectional survey metrics instead of tracked actions |
| Pal et al. (2021) | Users of financial applications through a structural model | Attitude and planning undergo moderation, leaving self-efficacy unaffected | Outcomes diverge somewhat from prior research |
Notifications do not affect everyone in the same way
The identical study also features a finding concerning alerts that warrants a dedicated section, since it runs counter to the arguments of both proponents and detractors: price trend notifications improve the learning of investors with correct beliefs, but reinforce the trading mistakes of those whose beliefs are incorrect.
In other words, the same alert amplifies what the user already brought with them. For someone who understands the market it provides useful information; for someone with a wrong model it confirms the error and pushes them to act on it. A feature that is neutral by design produces opposite effects depending on the user.
How much nudges amplify risk
The same authors published a second experiment in Journal of Behavioral and Experimental Finance, centered around risk instead of volume. An aggregate of 605 individuals across four nations traded a virtual asset within an experimental platform, featuring a gamified interface complete with digital nudges—such as achievement badges and motivational messages—that were expressly created to promote holding behavior.
The outcome: nudges markedly boost risk-taking propensity, especially within high-volatility settings.
And the variable that modulates the effect is the same one that appeared in the earlier study, now quantified: the effect is more pronounced among inexperienced traders and those with lower financial literacy, and a one-standard-deviation increase in financial literacy reduces the impact by 56 %.
| Factor | Effect on behaviour | Practical reading |
|---|---|---|
| Hedonic gamification | +5.17 % of trading volume | Real, but modest against the usual headline |
| Platform self-selection | 70 % of the observed difference | Comparing users of two apps is misleading |
| Nudges to hold positions | Amplify risk-taking | Worse the higher the volatility |
| Financial literacy | −56 % of the impact per standard deviation | The best measured protective factor |
| Trend notifications | Ambivalent: they teach or reinforce the error | Depend on the user’s prior beliefs |
The study with two portfolios at once
Duygu Şenol and Ceylan Onay devised an experiment that circumvents a frequent issue within this domain: participants being aware of observation. The pair utilized transaction records from investors maintaining simultaneous active portfolios through an investment brokerage alongside a stock exchange simulation game, yielding a framework devoid of both observer and subject expectancy biases.
Their initial discovery holds significant methodological weight: investors display distinct prejudices across actual and virtual settings. Their behavior shifts fundamentally when genuine funds are at stake.
And the second contradicts the usual narrative about gamification: participating in a stock game affects all biases differently depending on the degree of participation. Overconfidence and the disposition effect can be mitigated and decrease with more active participation in the game, while familiarity and status quo biases increase.
The authors add who takes part, a figure that returns to self-selection: young, inexperienced investors with medium-sized portfolios, and men, are more likely to participate.
Which part of behaviour it acts on
Two studies utilizing structural equation models have sought to identify which specific link in the chain gamification actually influences. Their findings do not entirely align, and this divergence offers valuable insights.
Garima Agrawal and Gunjan Sharma, with data from 557 users of financial applications in India, find that gamification significantly moderates the relationships between financial planning and financial conduct, and between financial self-efficacy and conduct, but not between financial attitude and conduct. And they add what experienced users ask for: real-time feedback, progress tracking and social sharing.
Abhinav Pal and two co-authors, with an analogous model, find a partly different pattern: gamification features significantly moderate the influence of financial attitude and of financial planning activity on financial management behaviour, but not the influence of self-efficacy.
What both have in common, and what deserves to be preserved: gamification does not impact behaviour directly, but instead by magnifying or reducing the influence of inclinations the user already possessed. It functions as a multiplier, rather than a motor.
Practical steps to apply this
Ordered by the strength of the evidence behind each point:
- Do not compare users of two different applications. 70 % of the activity difference is self-selection, not design.
- Turn off trend notifications if you have no thesis of your own. They reinforce the mistakes of users who arrive with incorrect beliefs.
• Expect the effect to grow with volatility. Nudges amplify the risk taken above all in volatile environments.
• Financial literacy is the measured buffer. One standard deviation reduces the impact of nudges by 56 %.
• Do not assume a simulator is either harmless or educational. It lowers overconfidence and the disposition effect, and raises familiarity and status quo bias.
Frequently asked questions
Are gamified trading apps driving higher user activity?
Somewhat, but less than the direct comparison suggests. In the randomised experiment published in Management Science, hedonic gamification increases trading volume by 5.17 % on average, but the difference in activity between gamified and non-gamified platforms is 70 % due to self-selection and only 30 % to gamification.
Who do platforms with confetti and badges attract?
A specific profile, and that is the problem. Participants with lower financial literacy prefer platforms with hedonic gamification elements, and those who prefer that gamification display noisy trading strategies, while those who prefer non-gamified platforms show more contrarian behaviour.
Do notifications help or harm?
Both, depending on who receives them. The same experiment documents that price trend notifications improve the learning of investors with correct beliefs, but reinforce the trading mistakes of those whose beliefs are incorrect. The feature has no single effect: it depends on the user’s mental model.
Do they make people take more risk?
Yes, and it is measured. In the experiment with 605 participants from four countries, digital nudges designed to encourage holding positions significantly amplify risk-taking, particularly in high-volatility environments, with a more pronounced effect among inexperienced traders and those with lower financial literacy; a one-standard-deviation increase in financial literacy reduces the impact by 56 %.
Does a stock market simulation harm a trader’s performance?
Not uniformly. In the study of investors who held a real portfolio and a game portfolio at the same time, overconfidence and the disposition effect are mitigated and decrease with more active participation in the game, while familiarity and status quo biases increase, and investors show different biases in the real and the simulated environment.
What we do not know
- What happens with real money. The available causal measurements are experiments with virtual assets, not with client portfolios.
• How long the effect lasts. The experiments are short sessions; habituation to the stimuli over months is unmeasured.
• Whether the net balance is positive or negative. Greater volume and higher risk do not inherently equate to an inferior result, yet scarce studies assess ultimate returns.
• Where the boundary of acceptable design lies. The distinction between a useful nudge and a manipulative one is much discussed and little operationalised.
References
1. Chapkovski, P., Khapko, M. and Zoican, M. Trading Gamification and Investor Behavior. Management Science, 2024. DOI: 10.1287/mnsc.2022.02650
2. Chapkovski, P., Khapko, M. and Zoican, M. Gamified risk-taking. Journal of Behavioral and Experimental Finance, 2025. DOI: 10.1016/j.jbef.2025.101049
3. Şenol, D. and Onay, C. Impact of gamification on mitigating behavioral biases of investors. Journal of Behavioral and Experimental Finance, 2022. DOI: 10.1016/j.jbef.2022.100772
4. Agrawal, G. and Sharma, G. An Evaluation of Gamification on Financial Conduct in Relation to the Effectiveness of Behavioral Traits. Financial Planning Review, 2025. DOI: 10.1002/cfp2.70016
5. Pal, A., Indapurkar, K. and Gupta, K. Gamification of financial applications and financial behavior of young investors. Young Consumers, 2021. DOI: 10.1108/yc-10-2020-1240
This content is for information only and does not constitute financial advice or an investment recommendation. The studies cited analyse experimental settings and specific samples that are not comparable with one another, and none of them allows the effect of any particular platform on any particular user to be anticipated.
