Privacy and trust-related concerns have been raised by the public concerning how COVID-19 health and tech stakeholders will handle users’ privacy and data [7]. For example, most Americans may trust COVID-19 stakeholders such as public health agencies and universities, but they do not trust tech companies such as Apple and Google, which developed the privacy-preserving Google/Apple Exposure Notification system, which most of the contact tracing apps on the market require and support to function properly [12]. A cross-section of US smartphone users was asked the question, “How much, if at all, do you trust _____ to ensure that people who report being diagnosed with coronavirus using their smartphone app remain anonymous — a great deal, a good amount, not too much or not at all?” A total of 56% of those polled (ie, nearly 3 in 5) did not trust tech companies such as Apple and Google, but 57% and 56% trusted public health agencies and universities a great deal or a good amount, respectively [23]. The limited trust in tech companies such as Apple and Google (<45%) may not come as a surprise given the widely reported Facebook-Cambridge Analytica Scandal about the 2016 United States elections [24].

High uptake is crucial for exposure notification apps to be effective in mitigating the spread of COVID-19. However, according to Walrave et al [25], “it remains unclear how we can motivate citizens to use these apps.” Although the government and tech companies have taken some measures to increase public trust by way of decentralization of collected data [12], Bluetooth contact tracing, and nontracking/storage of users’ location data via global positioning technology, much is yet to be done in the area of persuasive design to increase the adoption rate. For example, the current version of the COVID Alert app is minimalist [26] and lacks motivational affordances and incentives [27]. Motivational affordances are the persuasive elements that satisfy users’ needs. According to Zhang [28], when an information and communication technology (ICT) satisfies users’ motivational needs, they feel enjoyment and want it more. Hence, “the ultimate goal of designing an ICT for human use is to achieve high motivational affordance so that users would be attracted to it, really want to use it, and cannot live without it” [28]. However, “[a]part from providing receiving notifications about possible infections, current contract tracing apps appear to not provide a clear benefit to the user” [29]. Specifically, most of them lack vital persuasive features that motivate people to use digital health technologies to monitor and manage their health behaviors. Hence, the lack of persuasive features may contribute to low adoption rates of many contact tracing and exposure notification apps on the market [30].

Digital health researchers have stated that incorporating persuasive features into contact tracing apps could increase their adoption and use by the wider population [27]. In other words, contact tracing apps are more likely to be effective as persuasive technologies than as traditional information systems focused on functionality.

Persuasive technology is an interactive system intentionally designed to change attitudes or behaviors positively through persuasion and social influence but not through coercion or deception [31]. However, the current version of the COVID Alert app lacks basic persuasive and social influence principles that can motivate more users to download and use the app more frequently. Figure 3 shows the three main functional UIs of the COVID Alert app: “No Exposure,” “Exposure,” and “Diagnosis Report.” Apart from being minimalistic, all three UIs do not support essential persuasive features such as monitoring of the users’ daily contacts and exposure time. This may help them regulate themselves concerning observing social (physical) distancing guidelines in public settings.

There are two main design frameworks commonly used in designing and evaluating persuasive technologies. The first framework is called Cialdini’s [33] principles of persuasion, which comprise six persuasive techniques: authority, commitment, reciprocity, liking, consensus, and scarcity [34,35]. The second framework is called the persuasive system design model [36], which comprises 28 persuasive techniques and extends Fogg’s [31] seven persuasive techniques. The persuasive system design model includes four broad categories (primary task support, dialogue support, system credibility support, and social support) as shown in Figure 4 [36,37].

First, primary task support includes persuasive techniques that help the user to carry out the target behavior easily and effectively. Second, dialogue support includes persuasive techniques that motivate the user to perform the target behavior through feedback and interaction with the persuasive application. Third, social support includes persuasive techniques that motivate the user to carry out the target behavior through social influence. Finally, system credibility support includes persuasive techniques that make the persuasive application look credible to the user [38].

Each of the four categories in the persuasive system design model comprises seven persuasive techniques. Figure 5 shows three persuasive techniques in each of the four categories relevant to contact tracing apps. For example, primary task support comprises self-monitoring, tailoring, and personalization, and social support includes social learning, social comparison, and normative influence. These techniques, widely studied in persuasive technology research, have proven effective in changing health behaviors such as physical activity [39,40]. Moreover, dialogue support comprises praise, reward, and feedback. In particular, reward, be it virtual, tangible, or monetary, holds potential in motivating behavior change, as people from both high-income and low-income countries are receptive to it [41]. Finally, credibility support comprises trustworthiness, surface credibility, and authority. Research [36] shows that persuasive apps perceived as trustworthy and credible are more likely to motivate behavior change. Prior studies found a direct or indirect relationship between source trustworthiness [42] or perceived credibility [43] and behavioral intentions. Moreover, Oyibo et al [44] found that people from both high-income and low-income countries are receptive to the authority strategy. Interestingly, current exposure notification apps on the market are already equipped with the authority and credibility strategies by default given that they were sponsored by national governments that symbolize authority. However, the issue of trust in the area of data protection and privacy remains a roadblock to adoption [23].

Persuasive techniques are implemented in most mobile health apps on the market to motivate behavior change and help users achieve their goals. Figure 6 shows a fitness app called “BEN’FIT,” in which reward/self-monitoring and social learning/social comparison are, respectively, implemented in the personal and social versions (Oyibo et al [45]). Self-monitoring enables the user to track their physical activity, including calories burned and step count over time. Regarded as the cornerstone of persuasive apps, self-monitoring fosters self-awareness and commitment, among other advantages shown in Figure 7 [46]. In the context of contact tracing apps, Cruz et al [47] found that over 50% of their surveyed participants wanted to know how many infected people they have come in contact with and how many infected people have passed through a given location. Reward provides users with something to strive for and reinforces behaviors [48]. Feedback allows the user to get important information about their behavior at specific points in time, for example, after achieving a 10,000 steps milestone. Feedback is not listed as a dialogue support feature in the persuasive system design model, yet it is used as a persuasive feature in motivating behavioral change. Social learning and social comparison, which are correlated [49], use social pressure to motivate the target behavior [48].

In the no exposure UI, a self-monitoring feature, which tracks daily contacts and exposure time, and showcases historical behavior, can be incorporated in the second half of the screen, which is currently blank. The implementation of the self-monitoring feature is presented in Oyibo et al [51]. In the social version, a social comparison feature, which compares the user’s exposure levels (daily contacts and exposure time) with those of others in the community, can be incorporated as well. In addition, users can be allowed to customize the app (eg, choose a happy face avatar instead of a green hand icon that represents their no exposure state). Research shows that well-designed avatars can improve the user experience by drawing a closer connection between the user’s lived and digital identities as, for example, avatars possess some human signifiers like facial expressions that convey emotion [52]. This is in line with the liking principle in the persuasive system design model (Figure 4), which states that people are more likely to be persuaded by people similar to them or that are attractive [33,36].

In the exposure UI, a self-monitoring feature, which tracks the total number of contacts and approximately when the user was exposed, can be incorporated in the middle of the screen, as shown in Figure 8. The implementation of the self-monitoring feature is presented in Oyibo et al [51]. As in the no exposure UI, users should be able to customize the app (eg, choose a sad face avatar instead of a purple hand icon to represent their exposed state). In addition, in the social version, they should be given the choice to compare their exposure levels with those of others in the community as an additional means of motivation and insight.

In the diagnosis report UI, a social learning feature, which informs the user about the number of persons that have reported their COVID-19 diagnosis for a given period (eg, day or week), can be incorporated in the middle of the screen as shown in Figure 8. This additional statistical information can encourage users, when infected, to report their diagnosis to ensure the safety of the community. The implementation of the social learning feature is presented in Oyibo et al [51]. Moreover, users can be praised or rewarded for reporting their diagnosis. In a recent study, Jonker et al [53] found that respondents preferred apps that offer them incentives such as a token monetary reward (€5 [US $6] or €10 [US $12] a month), permission to gather in small groups (eg, after recovering), or free testing for COVID-19 after receiving an exposure alert.

In addition to the 12 persuasive features drawn from the persuasive system design model (Figure 4), hot spot monitoring, which we call “Social Location Monitoring,” can be used as a persuasive strategy to promote adoption and use. Social location monitoring is the tracking and gathering of information about a location that includes the number of persons who are infected that currently reside in, have been to, or passed the location in a given period to help users make informed decisions. Figure 8 shows a hypothetical interface for incorporating social location monitoring to motivate beneficial behaviors (eg, avoiding hot spots, social distancing, and wearing a mask). In a recent study, Li et al [54] found that respondents were more willing to install contact tracing apps that collect users’ location data than those that do not, due to the additional benefits they provide about hot spot information and analysis. Social location monitoring can help local authorities allocate resources in a better way and enact better health care policies during the COVID-19 pandemic [55].