From freemium to premium: The conversion capability of digital nudges

© 2021 Published by VŠB-TU Ostrava. All rights reserved. ER-CEREI, Volume 22: 35–44 (2021). ISSN 1212-3951 (Print), 1805-9481 (Online) From freemium to premium: The conversion capability of digital nudges Zsófia GYULAI * Institute of Business Studies– University of Szeged, Kálvária av. 1., Szeged 6722, Hungary Abstract Companies attempt to sway consumers’ decisions at various points in the buying process. Their so-called “nudges” are one of the effective techniques that they use to influence the process. Nudges subtly guide the decision maker towards a predetermined outcome while preserving the freedom of choice. The aim of this study is to classify digital nudges and investigate their effect on conversion. During the investigation, the following approach was used to analyse web analytics data from an SaaS provider. First, conversion data were obtained and com-pared with those from the year before and the year after the use of digital nudges. Then, in the two periods, the number of upgrades – from the freemium to the premium package – was assessed, taking into account whether the number of upgrades came directly from clicking on a nudge. Digital nudges help to improve conversion rates while maintaining customers’ “normal” proclivity for page leaving. Loss aversion nudges were found to have the greatest tendency to convert throughout the study. Keywords Customer journey, digital nudges, software-as-a-service, conversion rate, website analytics JEL Classification: M31, M21 [email protected]-szeged.hu Ekonomická revue – Central European Review of Economic Issues 24, 2021 36 From freemium to premium: The conversion capability of digital nudges Zsófia GYULAI 1. Introduction End-user spending on software-as-a-service (SaaS) worldwide has tripled in the last five years. The increasing competition is forcing SaaS providers to stand out from their competitors (among other things) by designing an appropriate, experience-rich customer journey (Lemon and Verhoef, 2016). Due to its operational logic, most of the time, users encounter software-as-aservice on websites while using the software (Liao, 2010; Joha and Janssen, 2012). As a result, SaaS businesses need to pay special attention to the design of the web page (Batra, 2017). There is too much information available for customers to consider all the possible options; therefore, human decision making is irrational (Simon, 1955). Infor-mation dumping caused by digitalization has only increased further the amount of information available (Baxendale, Macdonald and Wilson, 2015; Haan, Wiesel and Pauwels, 2016) . More and more sources of information make it even more difficult to make a rational decision (Winer and Margolis, 1983). Thus, companies want to intervene in the customer decisionmaking process to ensure that the decision that is the most favourable to them is made. A good way to achieve this is to use so-called nudges (Thaler and Sunstein, 2009). SaaS providers need to influence the decision-making process of their visitors and users through various digital nudges while maintaining the level of experience of their customer journey (Thaler and Sunstein, 2009; Batra, 2017). In this study, I search for the answer to the following research question: What is the impact of each digital nudge on conversion? A further aim of the study is to explore the related web analytics effects. 2. Literature Review 2.1 The customer journey of a software-as-a-service Before processing the relevant literature, it is essential to clarify the concept of Software-as-a-Services. The Software-as-a-Service (SaaS) model is a modern way of providing software services that are based on the idea of offering software as a service rather than a product (Bennett et al., 2001). Customers are given a single set of functionalities with minimal options for customer-specific changes, and vendors strive for economies of scale (Batra, 2017). Customer relationships are more straightforward and continuous in the digital product industry, pricing is focused on subscriptions, and both software creation and hosting are key activities and necessary capabilities (Luoma, 2014). In the Software-as-a-Service model, a centralized data center adopts an application or service through a network – the Internet, Intranet, LAN, or VPN – and charges a monthly fee for access and use (Mikhramova, 2010). The most common Software-asa-Services are in the field of invoicing, customer relationship management, and content management, but SaaS-s for teamwork and analysis of website analytics data is also popular. For the B2C sector, also available Software-as-a-Services for listening to music, watching videos, and making creative materials (Luoma, 2014). With the growth of the service economy, service providers' ability to offer customer-centric services has become increasingly important. Providing high-quality experiences to consumers is a long-term competitive advantage with a strong financial effect for businesses. (Halvorsrud, Kvale and Følstad, 2016). As a result, it's critical for SaaS providers as well to have an exceptional customer experience, which can be accomplished, among other things, by tracking customer’s journey in real-time. Customer experience and customer journey are two inseparable concepts in online consumer behavior research. User experience refers to customers’ perceptions of the brand in their relationships with the brand, which will be the combined result of all interactions (McColl-Kennedy et al., 2015). These perceptions emerge from the customer journey, the touch points between the brand and the user, and the quality of the user’s environment (e.g., speed of the Internet, condition of the electronic device used) (Meyer - Schwager, 2007; Lemon and Verhoef, 2016). The user experience is generally considered holistic, encompassing the customer’s cognitive, emotional, social, and physical responses to a provider, brand, or product in direct or indirect relationships that develop through multiple touch points throughout the customer journey Z. Gyulai - From freemium to premium: The conversion capability of digital nudges 37 (McColl-Kennedy et al., 2015; Lemon and Verhoef, 2016). The customer experience encompasses all aspects of a company’s offering — customer service quality, advertising, packaging, product and service features, ease of use, and reliability. A good customer experience means that every user encounter with the brand is as expected (Lemon and Verhoef, 2016). Lemon and Verhoef (2016) defined customer journey as a customer's set of interactions with a firm through various touch points during the purchase cycle. Existing research suggests that multiple touch points can be found during the customer journey (Meyer, Christopher; Schwager, 2007; Baxendale, Macdonald and Wilson, 2015). In each stage of the experience, the customer can interact with each of these touch point categories. The intensity or value of each touch point category may vary depending on the nature of the product/service or the customer's journey at each level. Firms must then determine how key touch points can be influenced once they have been identified (Lemon and Verhoef, 2016). In the case of software-as-services, the user and the company meet most often through the website where the user accesses the particular service. This is why SaaS companies should pay special attention to their website (Batra, 2017). In this research, we considered the website as a company-user touch point and we examined the impact of digital nudges appearing on the website on conversion. 2.2 Business models of software-as-a-services Laying the theoretical foundations of business models has started to emerge only in the last decade (Barakonyi, 2008). A business model explains how an enterprise generates, distributes, and captures value (Osterwalder and Pigneur, 2010). Digital business models vary from traditional business models in that they make use of technical advances to conduct business and produce revenue in new ways (Joha and Janssen, 2012). The advancement of technology and the widespread use of the internet has resulted in a change in digital design, as well as advancements in software development technology and hardware efficiency, which facilitates software business model innovation and consumer segment changes (Liao, 2010). To understand SaaS business models, it is essential to examine the development of the SaaS operating principle and compare it with traditional software services. The software-as-a-service model evolved from the Application Service Provider (ASP) model, SaaS practically is an extension of the ASP model. Although ASP-s try to concentrate on handling and hosting thirdparty ISV (independent software vendor) applications, SaaS vendors handle their software. ASP-s often provide more conventional client-server programs, which require users to install software on their computers. SaaS, on the other hand, is entirely software-based and can be accessed via a web browser (Liao, 2010; Joha and Janssen, 2012). Traditional software services are called “commercial off-the-shelf” (COTS) software services in the literature. Users buy bundled software applications from the COTS provider for a one-time fee. To ensure successful integration with the user's current IT infrastructure, the application's source code is updated to meet the user's unique business needs. The provider bears an operating expense and is paid by the customer with a one-time charge. COTS users must have their own inhouse IT services to manage and operate the app. Under the SaaS model, the provider bears an initial setup cost to start the business with a customer, as well as a service cost per transaction to support the user. Users are charged a fee per transaction (Dan, 2007; Popp, 2011). Software-as-a-Services tend to use two main business models: one is a freemium subscription offer, the other is a free trial, but these can be applied at the same time. In the freemium model, basic services are provided for free, while premium services are available for a fee (Osterwalder and Pigneur, 2010). In contrast, the free trial model services are provided with full function but just for a limited time (Wang et al., 2013). Marketing professionals at SaaS companies use various digital nudges to make the users upgrade to the premium plan from the freemium plan and to subscribe after the free trial (Koch, 2017). 2.3 Digital nudges research By researching choice architecture, or the sense in which people make choices, psychologists have studied how to convince humans to follow new behaviors or ideas. Nudge theory is one of many facets of choice architecture that can influence human actions (Brown, 2019), and it is the one we want to concentrate on. According to conventional economic theory, human deed is often rational. In contrast to this, the psychological concept “nudge” claims, that humans can only behave rationally within such limits - due to cognitive limitations (Simon, 1955; Thaler and Sunstein, 2009). Thaler and Sunstein (2008) used first the term "nudge" and they defined it as any feature of the choice architecture that predictably changes people's actions without prohibiting any choices or substantially altering their economic incentives. The majority of nudging research has taken place offline mainly in the field of health care (Prainsack, 2020), politics (Mols et al., 2014), sociology (Room, 2016), or behavioral economics (Thaler and Sunstein, 2009), and digital nudging is still in its infancy. Ekonomická revue – Central European Review of Economic Issues 24, 2021 38 Nowadays, more and more people make decisions online, such as buying, booking vacations or other services, or even signing contracts, so digital nudging is also becoming increasingly important (Demarque et al., 2015; Eigenbrod, Janson and Leimeister, 2018). Digital nudging is the use of technology to build nudges that change human behavior in digital environments (Brown, 2019). Mirsch, Lehrer, and Jung (2017) identified twenty different psychological effects in their literature search related to nudging. 2.4 Types of digital nudges Among the psychological effects, those that may occur on the post-login SaaS interface have been presented as this is the subject of our research. We demonstrate each psychological effect through our examples as well as how they can occur in an online environment. Online companies often use framing theory as a psychological effect to create digital nudges (Eigenbrod, Janson and Leimeister, 2018). The central concept of framing theory is that a problem can be presented from several viewpoints and interpreted as having consequences for various values or considerations. People create a specific conceptualization of an issue or reorient their thought about an issue through the process of framing (Chong and Druckman, 2007). Nudges with framing psychological effect take advantage of the fact that people respond differently to different formulations of the same decision situation (Tversky and Kahneman, 1985; Chong and Druckman, 2007; Szántó and Dudás, 2017). Framing is the process of creating a decision frame in which the decision-makers' conception of the actions, consequences, and contingencies associated with a specific choice is driven by psychological concepts. Shifts and decision results are more predictable as a result, and odds are changed. A controlled presentation of a decision problem using various framing techniques for one decision problem is referred to as framing (Tversky and Kahneman, 1985). In the example of Tversky and Kahneman (1985), if a person has already lost $ 140 in a race bet and is considering another bet of $ 10, he will likely frame the last bet as the last chance to win. People who do not establish a reference point for their loss (maximum loss) can expect high losses. In the digital space, a good example of this is achieving the free shipping value of webshops. Buyers need to decide how much more they are willing to spend in the webshop than planned to avoid shipping costs. To achieve the amount of free shipping cost, webshops tend to remind the customer at several points in the customer’s journey. According to the example of Mirsch, Lehrer, and Jung (2017) webshops (such as Amazon) intervene in the choice architecture by placing ancillary or similar products on product pages, which can lead to additional, unplanned purchases. The psychological effect of loss aversion can be explained by the framing effect (Chong and Druckman, 2007; Szántó and Dudás, 2017). People respond differently to different formulations of the same decision situation: the same outcomes appear to be a benefit from some perspectives and a loss from others, but experience shows that we are much more susceptible to losses than gains of the same magnitude (Szántó and Dudás, 2017; Eigenbrod, Janson and Leimeister, 2018). Thaler and Sunstein (2009) explain the psychological phenomenon with an energy-saving campaign. Two campaigns were compared: one focused on profit (you can save $ 350 a year with energy-saving methods) and the other on loss (you can lose $ 350 a year if you don't use energy-saving methods). The campaign emphasizing the loss had a better impact on people. Mirsch, Lehrer, and Jung (2017) cite the Booking.com website as an online example, where the information page of each accommodation indicates how many people are currently viewing the accommodation. With this nudge, Booking.com suggests to accommodation seekers that they will miss out on an offer if they don’t book, as many people look at the same offer. For Sofware-as-a-services, a good example of taking advantage of loss aversion is that the descriptions of each plan are not worded in terms of what they can gain from the subscription, but what they can lose if they do not subscribe like “Here’s what you are missing out”. The status quo bias explains people's deep desire to stick with the status quo because the risks of changing outweigh the benefits of staying the same, therefore, choosing a starting position is crucial for companies (Mirsch, Lehrer, and Jung 2017; Samuelson and Zeckhauser 1988; Szántó and Dudás 2017; Thaler and Sunstein 2009). In their research, Giesen et al. (2013) asked participants to choose a hamburger from a menu. Participants have divided into three groups: the first group had a large portion of french fries as the default option, the second had a small portion, and the third had no default option. Research results showed that participants chose the default garnish more often. For Software-as-a-Service providers, this pattern of behavior is useful for subscription plans. According to research, if a larger, more expensive subscription plan is set as the default option, users are more likely to choose it. Webshop owners can prefer this behavior by selecting package insurance by default or by setting their preferred payment option. Mirsch, Lehrer, and Jung (2017) cite Tesla’s online configuration tool as an online example of the status quo bias. When configuring models, the website offers various default options that typically steer the customer in the direction of a more expensive cart value. According to the example of Thaler and Sunstein (2009), it is due to the status Z. Gyulai - From freemium to premium: The conversion capability of digital nudges 39 quo bias that people do not cancel newspaper subscriptions even if they no longer read that newspaper. SaaS companies take advantage of this to develop each plan so that the subscription is automatically renewed until you cancel, usually monthly, semi-annually, or annually. Seeing the success of this, the FMCG sector is already applying this strategy, e.g. Function of Beauty, where you can purchase a 3-month hair care subscription, or Tokyo Treat, where you can subscribe to monthly sweets packs. The emotional association-based nudges build on the fact that novel, person-relevant, or spectacular influences can evoke emotional associations in individuals that significantly influence their decisions (e.g., deterrent images on tobacco product packaging) with other nudges, e.g. with framing nudges (Szántó and Dudás, 2017). A good example of nudges that use emotional associations in an online environment is the owl mascot of the Duo Lingo language learning app. For language learning, you can collect points from which you can buy clothes for the owl. Another good example is the squirrel mascot appearing on the inner surface of the Ingatlannet.hu real estate portal, which is happy and rich when the real estate agent has bought enough credits, however, when the credits run out, the squirrel becomes sad and his treasure chest is emptied. Individuals may plan for a scenario in which they must make a decision. Specific subjects, moods, questions, or facts may be added before a decision is taken, for example, by visualizing the implications of a decision (Tversky and Kahneman, 1985; Friis et al., 2017). Priming is based on this psychological effect, which companies prefer to use when designing nudges (Thaler and Sunstein, 2009; Mirsch, Lehrer and Jung, 2017; Eigenbrod, Janson and Leimeister, 2018; Battaglio et al., 2019). A study by Blumenthal and Turnipseed (2011) highlights that the location of the poll can affect the outcome of voting: the pace of the voting booth evoked conservative values from voters. In the online space, the design of websites or social media pages can also affect the conversion outcome. In the example of Mirsch, Lehrer, and Jung (2017) they examined Instagram images of Air France. Images of the destinations contributed to the purchase of the flight ticket as the images steered the visitors towards the decision to travel. Focusing and emphasizing the attention of decision-makers on different information can lead to a change in the decision-making process (Saghai, 2013; Szántó and Dudás, 2017). For example, if the level of tax on alcohol products is already indicated on the price tag, it has a negative effect on alcohol consumption (Szántó and Dudás, 2017). As an online example, various pop-ups and notifications can be mentioned that can influence the visitor during the online buying decision process. 3. Methodology and Data The software we examined is an integrated web analytics software: it provides both quantitative and qualitative data to website owners. Although nudges can be displayed in many places by SaaS companies such as e.g. in the form of online advertisements, newsletters, or through social media (Mirsch, Lehrer and Jung, 2017; Eigenbrod, Janson and Leimeister, 2018), as SaaS users most often encounter the software on the website (while using the software), so we researched at this touch point. Many forms of appearance are also possible within the website, of which those relevant to our research have been discussed in the literature review. Within the website, we only looked at post-login nudges that appeared to users of the free plan and encouraged them to subscribe to the premium plan. These nudges provided a direct link to the premium subscription, so we were able to accurately measure its direct effects using Google Analytics. However, it is important to mention that nudges also have an indirect effect: regardless of whether the user did not click on the call-to-action of the given nudge during the session, it may have influenced it and may influence its subsequent decisions (Szántó and Dudás, 2017). 4. Empirical Results Our research aimed to examine the impact of digital nudges on the customer journey of the Software-as-aService. The nudges examined draw attention to the limitations of the free package: some features are not available or some data download volumes are limited. In the research, the type of restriction was not included in the study. On the one hand, the interpretation of nudges does not include relevant economic incentives (Thaler and Sunstein, 2009) (in this case, price and features), and on the other hand, these restrictions were already present in the base period. Nudges have been included on the site since July 2019, so the period between July 2019 and July 2020 is the reference period for the research, and the year before is the base period. Our first hypothesis is that (H1) digital nudges increase the conversion rate. To test the hypothesis, we used a secondary analysis of Google Analytics statistics. On the website, each nudge is marked as an event, so the web analytics software makes the conversions visible. Our second hypothesis is that (H2) digital nudges do not increase the exit rate on the examined website. We also used the web analytics statistical data for this part of the research: I examined the exit rate of the subpages with nudge, so the proportion of the exit page of the tested subpage in the base and reference period. Based on our third hypothesis, (H3) nudges af- Ekonomická revue – Central European Review of Economic Issues 24, 2021 40 fecting emotions result in the highest conversion propensity, for which we examined the conversion data attributable to individual nudges of the current period using web analytics software. 5. The identification and categorization of the nudges on the website In the course of our research, we examined the nudges that appear on the dashboard while using the service, which encourages the user to upgrade the freemium plan to a premium subscription. The first nudge examined (Figure 1) is a notification at the bottom and top, which reminds us that the free version used only updates the number of sessions on the graph once a day. The notification below also draws attention to the user’s ability to work more efficiently by subscribing to the service. This nudge falls into the category of loss aversion, as the omission of subscriptions in communication appears as a negative, to be avoided, while subscriptions make work efficient. Figure 1 The first nudge examined Source: observational research Even in the base period, the graphs of quantitative data on the dashboard were updated only once a day, however, the company did not draw attention to this and did not offer a subscription option that would have resulted in a more frequent update. On the subscription page, there is also a loss aversion nudge (Figure 2) where the phrase “Here’s what you’re missing out” takes advantage of the psychological impact of loss avoidance. Figure 2 The second nudge examined Source: observational research On the date picker panel shown in Figure 3, the company has placed an eye-catching nudge that draws attention to the fact that the user can only access web analytics data from the last 30 days through the software. Due to the use of the word “only”, this notification can be considered a framing nudge. Figure 3 The third nudge examined Source: observational research In the base period, only 30 days of data were also available, but this was only perceived by the user as being unable to click on an earlier date in the date picker. The following two pop-ups fall into the category of focusing and emphasizing attention (Figures 4 and 5), however, emotions may also play a role in these nudges. Figure 4 The fourth nudge examined Z. Gyulai - From freemium to premium: The conversion capability of digital nudges 41 Source: observational research The pop-up in Figure 4 warns that the number of web pages that can be analyzed by the software has been exceeded, and the nudge in Figure 5 notifies that the user exceeded the viewing limit of session recordings. This nudge, with the surprised owl and barrier, can also trigger emotional associations that can also influence consumer decisions. During the base period, the website sent a notification to the user informing him that a certain limit had been exceeded, but did not offer the subscription option. Figure 5 The fifth nudge examined Source: observational research Also, the pop-ups in Figures 6 and 7 are intended to draw the attention of decision-makers to various pieces of information by obscuring the background. Pop-ups point out that the feature is not available as part of the free package. Figure 6. The sixth nudge examined Source: observational research Figure 7. The seventh nudge examined Source: observational research During the base period, these features were not available in the free packages, and the company did not display this option on the dashboard. Table 1 Nudge categories Nudge Appearance Nudge category 1. notifications on the dashboard loss aversion 2. subscription page loss aversion 3. date picker framing 4. pop-up with owl focusing and emphasizing the attention emotions 5. pop-up with owl focusing and emphasizing the attention emotions 6. a pop-up with blurred background focusing and emphasizing the attention 7. a pop-up with blurred background focusing and emphasizing the attention Source: observational research Table 1 summarizes the categories of nudges on the website and where they appear. 6. Findings and Discussion H1: Digital nudges increase conversion rates To test the hypothesis, I examined the aggregate results of the event tags belonging to the nudges, with the secondary dimension being the conversion, that is Ekonomická revue – Central European Review of Economic Issues 24, 2021 42 the subscription. I also compared the data of the web analytics software with the subscriber data from the enterprise resource planning system of the examined website. During the base period, 8% of free users subscribed to one of the packages, which increased to 12% during the period considered. 52% of conversions came from clicks on nudges. Based on the results, I accept my first hypothesis, so in the case of the examined website, the conversion rate increases due to the effect of digital nudges. H2: Digital nudges do not increase the exit rate on the examined page. The second hypothesis was tested for the first and second nudges, since these nudges are part of the page, the other nudges do not have their URLs, so the number of times there were exit pages cannot be examined on them either. To interpret the data, it is important to know that for the average user, the graphs on the home page of the dashboard are most open, and the exit page is usually this subpage. Another important circumstance about the research was that we were able to examine the exit pages at the URL level, however, the nudges appeared only to users with the free plan. The dashboard with the notification nudges was 2 547 times the exit page out of 3 252 visits during the base period, which means that 78.32% of the visits were the last visits of the site. During the period under review, the number of visitors to the dashboard increased to 3,836, but this sub-page served as an exit page 3002 times, representing 78.25%. The second nudge examined appears on the subscription page, which users are much less likely to visit. During the base period, this was the last page viewed by users 12 times out of 342 visits, representing 3.5% of visits. During the reference period, this ratio increased minimally to 3.8%. The analysis of the web analytical data showed that the exit rate on the examined pages did not increase significantly in the reference period compared to the base period, so we accepted our hypothesis. H3: Nudges that affect emotions result in the greatest willingness to convert To test the effectiveness of nudges, we used Google Analytics to list the event tags associated with the nudges. We examined the impressions of the event tags and the impressions that resulted in conversions during the reference period. Table 2 Nudge impressions and conversions during the reference period Nudge Impressions Unique events Conversions 1. 9 688 1 459 38 2. 211 192 2 3. 11 865 1 367 16 4. 98 96 0 5. 5 472 5 398 22 6. 587 562 8 7. 672 629 5 Source: Google Analytics We summarized the event impressions, the unique event impressions (event impressions by a single user), and the conversions calculated from the unique event impressions in Table 2. Table 3 Conversion rates of the nudge categories Nudge category Impressions Unique events Conversions Conversion rate loss aversion 9 899 1 651 40 2,4% framing 11 865 1 367 16 1,2% focusing and emphasizing the attention and emotions 6829 6685 35 0,5% focusing and emphasizing the attention without emotions 1259 1191 13 1,1% focusing and emphasizing the attention only with emotions 5570 5 494 22 0,4% Source: Google Analytics Table 3 shows that the highest conversion rate, calculated from unique event impressions, was achieved by loss aversion nudges, but this result was more affected Z. Gyulai - From freemium to premium: The conversion capability of digital nudges 43 by the first nudge, the notifications on the dashboard home page. Based on the results obtained during the analysis of the web analytics data, the third hypothesis was rejected since the conversion rate of nudges affecting emotions became the lowest of the examined nudges. 7. Conclusions and limitations Comparing the web analytics data of the base and reference period of Software-as-a-Service, we can see that the introduction of digital nudges resulted in a significant conversion rate increase. The number of conversions increased more significantly during the reference period than the number of times users clicked on the call-to-action on the nudge. Therefore, the study of the indirect effect of digital nudges (Szántó and Dudás, 2017) indicates a further research direction. There are various ethical concerns about nudges (Meske and Amojo, 2020), and digital nudges can be displayed by using tools (e.g., pop-ups and notifications) that can confuse the user (Mirsch, Lehrer and Jung, 2017). For this reason, we considered it important to examine whether the appearance of nudges increases the exit rate on the examined web pages. Based on web analytics data, there was no significant increase in the number of exits on pages on which the SaaS placed nudges. In examining the effect of each nudge category on conversion, loss aversion nudge resulted in the highest conversion rate. In this result, however, it is important to mention two limitations of the research. The first is that the interpretation of nudges does not include the relevant economic incentives (in this case, price and characteristics). Although feature limitations were present on the site during the base period, the SaaS did not draw attention to them. Because of this, the particular restriction has had an impact on users due to their individual preferences. The other limitation is the problem of classifying nudges. The categorization of individual nudges (e.g., nudges that affect emotions) can be subjective. We have not yet found a solution to this problem in the relevant literature. In our further research, we would like to examine with a series of qualitative research how SaaS users relate to digital nudges. References BARAKONYI, K. (2008) ‘Üzleti Modellek’, Vezetéstudomány, 39(5), 2–14. Available at: http://unipub.lib.uni-corvinus.hu/4010/1/vt2008n5p0214.pdf. BATRA, M. (2017) ‘Customer Experience-An Emerging Frontier in Customer Service Excellence’, American Society for Competitiveness, 15(1), 198–208. BATTAGLIO, R. P. ET AL. (2019) ‘Behavioral Public Administration ad fontes: A Synthesis of Research on Bounded Rationality, Cognitive Biases, and Nudging in Public Organizations’, Public Administration Review, 79(3), 304–320. doi: 10.1111/puar.12994. BAXENDALE, S., MACDONALD, E. K. AND WILSON, H. N. (2015) ‘The Impact of Different Touchpoints on Brand Consideration’, Journal of Retailing. New York University, 91(2), 235–253. doi: 10.1016/j.jretai.2014.12.008. BENNETT, K. ET AL. (2001) ‘An architectural model for service-based software with ultra rapid evolution’, in IEEE International Conference on Software Maintenance, ICSM, 292–300. doi: 10.1109/ICSM.2001.972742. BLUMENTHAL, J. A. AND TURNIPSEED, T. L. (2011) ‘he Polling Place Priming (PPP) Effect: Is Voting in Churches (Or Anywhere Else) Unconstitutional’, BUL Rev., 91, p. 561. BROWN, C. (2019) ‘Digital nudges for encouraging developer actions’, in Proceedings - 2019 IEEE/ACM 41st International Conference on Software Engineering: Companion, ICSE-Companion 2019. IEEE, 202–205. doi: 10.1109/ICSE-Companion.2019.00082. CHONG, D. AND DRUCKMAN, J. N. (2007) ‘Framing theory’, Annual Review of Political Science, 10(1), 103– 126. doi: 10.1146/annurev.polisci.10.072805.103054. DAN, M. (2007) ‘The business model of “Software-AsA-Service”’, in Proceedings - 2007 IEEE International Conference on Services Computing, SCC 2007, 701–702. doi: 10.1109/SCC.2007.118. DEMARQUE, C. ET AL. (2015) ‘Nudging sustainable consumption: The use of descriptive norms to promote a minority behavior in a realistic online shopping environment’, Journal of Environmental Psychology. Elsevier Ltd, 43, 166–174. doi: 10.1016/j.jenvp.2015.06.008. EIGENBROD, L., JANSON, A. AND LEIMEISTER, J. M. (2018) ‘How Digital Nudges Influence Consumers – The Role of Social and Privacy Nudges in Retargeting’, in Twenty-Sixth European Conference on Information Systems (ECIS2018), Portsmouth, UK, 2018, 1–14. doi: 10.5465/ambpp.2018.11298abstract. FRIIS, R. ET AL. (2017) ‘Comparison of three nudge interventions (priming, default option, and perceived variety) to promote vegetable consumption in a selfservice buffet setting’, PLoS ONE, 12(5), 1–16. doi: 10.1371/journal.pone.0176028. GIESEN, J. C. A. H. ET AL. (2013) ‘Changing the default. How to promote healthier food choices’, Appetite. Elsevier Ltd, 71(2013), p. 475. doi: 10.1016/j.appet.2013.06.026. HAAN, E., WIESEL, T. AND PAUWELS, K. (2016) ‘The effectiveness of different forms of online advertising for purchase conversion in a multiple-channel