What’s this all about?

The year 2020 has seen the Covid-19 virus accelerate the trend of purchasing major items online as opposed to in-store. But spending a significant sum of money on a product sight-unseen is a bit uncomfortable at best, and downright nerve-wracking at worst. So people rely on product reviews to guide them.

I set out to discover exactly how people utilize and consume review content as part of the process of making a major purchase, and to find out if there were any opportunities to make the experience better.

User interviews

To start, I crafted a script and a list of open-ended questions and conducted interviews with five Seattle-area individuals to get a broad picture of how different people utilize review content. No surprise, there were behaviors that were specific to each individual, but there were also three significant takeaways from behaviors that all users shared.

People pay some attention to review content for just about every purchase online.
People pay significantly more attention to review content for high-dollar purchases, often (but not always) comparing user reviews and using a search engine to find ‘expert’ reviews.
Although people depend on reviews, they are universally suspicious of both user and expert reviews.

It’s this last point that jumped out as the biggest area of opportunity. People are suspicious of user reviews, wondering if they’ve simply been written by a product’s manufacturer, or if a person perhaps received a free product or gift in exchange for a favorable review. This latter suspicion also applies to ‘expert’ reviews, meaning reviews written by experts in the field or simply professional review writers.

This understanding of a universal pain point, along with the other findings from my interviews, allowed me to start to map out the journey users go through when they make a significant purchase online.

User journey

Although the idea is that users considering major purchases of all kinds will go through a similar journey, I’ve chosen to use the example of a television purchase for several reasons; it’s still likely to be a major expense for many people, it’s likely to be used for many years, and reviews on televisions are plentiful, both in terms of user-generated and expert-generated reviews.

Looking at the journey map, it’s easy to see opportunities for improvement. Not only does skepticism breed anxiety, but if a person wants to be as informed as possible, there’s a lot of website-hopping required. I wanted to explore solutions that would make people feel empowered, informed, and confident in their purchases online. So I had an idea of where I wanted to go, but first, I wanted to look at what’s already out there; what people are seeing and experiencing as they go through this process.

Competitive tear down and task flows

I looked most closely at Consumer Reports Best Buy to get an idea of how users are presented with review content, and how those sites worked in terms of navigation. I was able to latch on to some aspects of these sites I wanted to incorporate, such as approachable language and simple scoring, but also some that I wanted to improve upon, such as navigation.

*A competitive teardown of core screens, and a task flow for researching a product purchase on Consumer Reports.*

Who Said What comes to life

Now that I had an idea of my goal (empowering and informing online shoppers) and how prominent sites present their review content, I was able to hone in on a solution: Who Said What.

Who Said What is a review aggregator. It crawls the web and finds user reviews and expert reviews from a variety of sites, averages their score and presents them in an accessible format so people can quickly get an idea of what the product is really like without having to hop from site to site on their own. As one of my classmates put it, it’s like Rotten Tomatoes but for major products instead of movies.

I then set about creating a task flow, seen above, which gave me insight into possible pain or failure points (a user exiting). It also helped me with some foundational ideas on what the basic navigation could be and some core functional needs of the app.

Sketching and wireframes

Now that I had a representative task flow for the core functions of Who Said What, it was time to put my cheap Bic pen to paper and make some sketches for how those screens would look to users. I utilized stencils and the basic design principles of contrast, repetition, alignment and proximity to optimally organize the information.

*These sketches show various ideas for a selected product page.*

From there, I started working in Adobe XD to get wireframes into place, with a focus on making sure that the information on the page was organized logically and wouldn’t overwhelm users.

Prototyping and user testing

I refined my wireframes until I had a medium-fidelity prototype created, with the purpose of putting it in front of some people to see what they made of it. I created a research plan to help guide me through the process and make sure I came away from the testing with some useful and actionable information.

The study goal was to explore the strengths and weaknesses of the app’s structure and presentation of information. I crafted the following research questions I hoped to have answered:

Can people identify the purpose of Who Said What upon launching the app?
Are people able to find reviews on a product they’re interested in?
Are people to then find reviews on another product they may be interested in without getting lost?
Are users able to parse the presentation of review content on hub and product pages of the app?
What additional information, if any, do users want or need when considering a major purchase?

Using a combination of in-person and virtual observation using video conferencing software, I conducted usability tests with three participants. These sessions were invaluable not only in terms of getting some takeaways for how to improve the app, but simply as a learning experience for me, as the facilitator of the sessions.

Tap here to load the prototype

*Shown here is the navigation flow on my medium fidelity prototype.*

In the future, I’d like to do usability studies of a mobile app on an actual mobile device instead of people’s laptops; some conventions can be confusing when people are navigating a touch interface with a mouse instead of their finger. I also found that the search bar was something everyone valued, but as it’s nigh impossible to prototype, users were a little put-off by not being able to use it. On the positive side, all users were able to ascertain the purpose of the app as well as navigate within the bounds of the prototype without issue.

Final thoughts

I will be planning to look more at UI design conventions to create a more polished, high-fidelity prototype to build on the medium-fidelity prototype created for this class and case study. Overall, I’m pleased with the idea and how it came together. I honestly think that such a product as Who Said What would be useful for a variety of users, though I’m not naïve enough to pretend that the back-end work of creating what is essentially a web crawler / search engine for review content will be easy.

In all, though, I can see this concept having the backing of consumer watchdog and advocacy groups as well as even manufacturers themselves. It’s an opportunity to build trust with the greater consumer public, and therefore a means to strengthen brand identity and inspire greater customer loyalty.

Research paper written as final project for Communication Leadership 517, The Psychology of User Experience, at the University of Washington

Introduction

Say you’re a novelist. The day has wound down, the sun is setting and you’ve found yourself on a streak. Words and sentences are pouring out of you like water, falling into a beautifully cohesive river of plotline. Then your phone buzzes in your pocket. It might be your something important. Someone saying hello, wondering how you are, or perhaps someone asking if dinner’s in the oven yet.

Dam. You stem the literary flow and check your phone, only to find a different type of notification. It’s a college friend’s birthday, or there’s a cold front expected tomorrow, or maybe you should redeem those credit card rewards, like, right now.

And after having seen one of those notifications, you tap it, read it, and about five minutes later, realize you’ve read a few more things and have totally lost track of what you were doing. Meanwhile, the fountain of phrases has gone dry. Might as well get dinner in the oven, then.

What are push notifications anyway?

Today, mobile push notifications are an effective and popular means of getting information in front of a diverse audience from a variety of sources, including e-mail, news, social media, e-commerce and more. They have expanded from a concept that was popularized by Research in Motion (RIM) and incorporated into the Blackberry 850 in 1999, wherein the device would continuously receive e-mail without the user having to request it (Britannica, 2013)i. Before this innovation, and indeed in earlier versions of the Microsoft Outlook e-mail client, users would have to manually check their e-mail to see whether they had received any messages – and this could become stressful if you were expecting an important one (Evans, 2017)ii.

Since the user didn’t know how long they would have to wait until the e-mail they were waiting for was to become available, they would often fall back on clicking the send/receive ‘refresh’ button far too often. This comprises what is known as a variable-interval schedule of reinforcement, where the user’s ‘reward’ comes on an unpredictable time scale (Ferster and Skinner, 1957)iii. Later, Outlook adopted the same type of real-time ‘push’ e-mail that RIM had pioneered with the Blackberry; all messages were automatically downloaded, so if you hadn’t received a notification about an e-mail, you could safely ignore e-mail altogether and go on about your day until what you were expecting was received.

But, eventually, you would receive an e-mail. And even though you didn’t have to click ‘refresh’ to find out if you had a new one, you might still find yourself anxiously waiting for an important reply. When it finally does arrive, well, ping goes the notification.

A study conducted in 2014 found that their participants had to deal with, on average, 63.5 notifications per day (MobileHCI, 2014)iv. A more recent study by Business of Apps, an app industry resource for news and analysis, pegs that number at 46 notifications per day (Business of Apps, 2019)v. The same industry resource claims that Android automatically enables push notifications for apps, while iOS devices (provided by Apple Inc.) do not. In any case, many of us, on any given day, are bombarded by push notifications in some form or other.

Why do people engage with push notifications?

Let’s return to that concept of variable-interval reinforcement schedule we mentioned earlier. Basically, services such as text-messaging, e-mail, breaking news, real-estate listings, sporting-event scores may reward users on an unpredictable time scale utilizing push notifications (Evans, 2017)ii.

Some apps give you control over what types of notifications you receive (left, Wall Street Journal); some only let you enable or disable them generally (right, Washington Post). Giving users control means they’re less likely to be overwhelmed them with notifications they don’t want.

Each push notification acts as a nudge, an ‘ahem,’ a tap on the shoulder. You don’t know when one is going to arrive, and when it does, you don’t know for sure if it’s something that will appeal to you. The fact remains, people interact with them, and then interact with their host apps afterwards.

But is this necessarily bad? After all, push notifications can bring something interesting to your attention to keep you entertained, particularly when you’re in a task-negative mindset. A task-negative mindset means you aren’t actively engaged in an effortful process (and will actively resist them), and are open to distractions (Gordon, B.A., Tse, C.Y., Grattaon, G. and Fabiani, M, 2014)vi. You can engage with content behind a push notification while walking (don’t fall into a fountain), riding the bus, or lounging on the couch.

However, if you are in a task-positive mindset, push notifications can be an unwelcome distraction, and can lead to switching costs. Switching costs come about because we aren’t capable of true multitasking as a result of our limited working memory (Cowan, 2005)vii. And switching costs are real; the likelihood of errors increases, and over the course of the day, a person might lose 40% of their productivity due to task switching (Weinschenk, 2012)viii.

Can’t we just put down the phone? Ignore the notifications? Or just address a single notification and move on? Today, many apps that serve up notifications are designed around variable-ratio schedules of reinforcement, which has the potential for an even more addictive effect than variable-interval schedules of reinforcement (Ferster and Skinner, 1957 (as cited in Evans, 2017))iii. Variable-ratio schedules occur when a user has no idea how many clicks or scrolls are required to find something worthwhile, and so they keep clicking, scrolling and exploring until they do.

A 2014 study using a simple weather forecast app brings some insights. There were two groups of participants; all participants regardless of group were required to open the app at least once a day. One group received no push notifications, while the other did. The study found that users in the group that received push notifications interacted more with the app than users in the group that did not receive them. Additionally, those who clicked on the notifications interacted with the app more than those who received them but did not click (Kim, 2014)ix.

Thus, there is evidence that push notifications can lead to interaction with apps that may utilize a variable-ratio schedule of reinforcement, which can then be especially difficult to discontinue using.

Proposed solutions to aid in self-regulation, and potential outcomes

The proposed design solution to this problem is a feature called ‘scheduled notifications.’ In an ideal world, a given app would prompt you upon install which notifications you want. Let’s use a news app as an example. The choices could be:

· No notifications

· Only breaking news items

· Choose which topics to receive notifications about

If a user has chosen either the second or third options above, the app would then prompt the user, “Would you like to schedule your notifications?” The user could then choose which time(s) of the day for notifications to come in. You could, for example, choose to get most of your news in the morning, or catch up on the day’s events in the evening. This is distinct from ‘Do Not Disturb’ or other OS-level features meant to limit screen time in that it would be easier for a user to set up on app-by-app basis, and it affords developers and businesses some interesting opportunities. Starting from the user’s perspective, if a user were to choose to schedule their notifications rather than receive them at random, there would be several distinct advantages.

First of all, this would transform the notifications from this app from a variable-interval reinforcement schedule into a truly fixed-interval reinforcement schedule. A user would know exactly when to expect to interact with the app, and therefore will have fewer prompts throughout the day tempting them to engage and then potentially get sucked in. Plus, if a user schedules appropriately, they are more likely to encounter notifications in a task-negative state, and will welcome them. Lastly, this would also help task-positive users from becoming distracted, introducing switching costs, or becoming annoyed with too many notifications and therefore producing a negative reaction to the app and brand behind it (Wohllebe, 2020)x.

This last point illustrates that not only would this be an ethical design choice for a company to make on behalf of its users, but it is also a potential business advantage as well. If you annoy a user with so many notifications that they disable them or uninstall the app, you’ve lost a customer. It’s better for business to have users around on their own schedules than not at all.

Plus, a daily download of notifications is something akin to a daily e-newsletter, only a business wouldn’t have to power through a user’s e-mail inbox, which is probably already bloated with spam, other newsletters, and actual important things that users already spend a lot of attentional capacity finding (Evans, 2017)ii.

A further bonus is that a business likely has a lot more flexibility in terms of design for revenue in an app environment than an HTML e-mail. You can have embedded video ads, auto-refreshing ads and more from within an app than simple, static ads in an e-mail. Of course, a further ethical consideration is resisting overloading a user with ads, which may also cause them to disable the feature.

One last feature possibility to come from scheduled notifications is that a developer could simply have a single scheduled notification, labeled as ‘Top Twenty’ or ‘Fab Forty’; for example, the top X number of Facebook posts, carefully curated from your closest friends, family and the pages you frequent most. A user could scroll through those and when they reach the end, something akin to the Instagram feature, ‘You’re all caught up!’ could interrupt the user and give them a chance to break the variable-ratio reinforcement schedule of endless scrolling and get on with their lives.

Conclusion

It’s true that this proposed design solution does require active opting-in on the part of the user, and so may not have as widespread an effect as if it were to be a standardized practice across all apps. But there are user interface and decision science design choices developers could make to help ‘nudge’ users into enabling the feature.

Evidence suggests that such a feature, particularly if implemented as a once-daily download resulting in fewer notifications overall, would have an overwhelmingly positive effect on users. A small but promising study recruited participants to turn off all phone notifications for 24 hours, and found that not only did participants who were anxious about doing so actually enjoy it, but two years later, some participants still had modified notifications settings as compared to the start of the study (Pielot and Rello, 2017)xi. And though testing would be required, there is reason to believe that such a feature could be good for business as well, particularly in terms of breaking through the ‘noise’ that we all encounter in our ever-more-connected lives.

Finally, giving the users a choice to take greater control over the information they see aligns somewhat with Nir Eyal’s philosophy on breaking smartphone addiction in his book Indistractible (admittedly his earlier book, Hooked, detailed strategies on fostering this addiction). He states that taking back control is slow and “it involves self-reflection. He argues that many times we look at phones because we are anxious and bad at being alone – and that’s not the phone’s fault” (Bowles, 2019)xii.

Scheduling notifications is a small step towards empowering users to take some control over their digital lives. Their attention doesn’t have to be at the mercy of businesses and app developers, and they don’t have to keep scrolling once they’ve finished with something a notification took them to.

References

i. Britannica, T. Editors of Encyclopaedia (2013, March 20). BlackBerry. Encyclopedia Britannica. https://www.britannica.com/technology/BlackBerry-wireless-device

ii. Evans, D. C. (2017). Bottlenecks: Aligning UX Design with user psychology. New York, NY: Apress.

iii. Evans, D. C. (2017). Bottlenecks: Aligning UX Design with user psychology. New York, NY: Apress.

Ferster, C. B., and Skinner, B.F. (1957). Schedules of Reinforcement. East Norwalk, Ct., US: Appleton-Century-Crofts.

iv. Martin Pielot, Karen Church, and Rodrigo de Oliveira. 2014. An in-situ study of mobile phone notifications. In proceedings of the 16th international conference on Human-computer interaction with mobile devices & services, MobileHCI '14. Association for Computing Machinery, New York, NY, USA, 233–242. https://doi.org/10.1145/2628363.2628364

v. Business of apps. Editors of business of apps (2019, November 6). Key Push Notification Statistics. https://www.businessofapps.com/marketplace/push-notifications/research/push-notifications-statistics/

vi. Evans, D. C. (2017). Bottlenecks: Aligning UX Design with user psychology. New York, NY: Apress.

Gordon, B.A., Tse, C.Y., Grattaon, G. and Fabiani, M (2014). Spread of activation nand deactivation in the brain: Does age matter? Frontiers in Aging Neuroscience, 6, 288.

vii. Evans, D. C. (2017). Bottlenecks: Aligning UX Design with user psychology. New York, NY: Apress.

Cowan N. (2005). Working Memory Capacity. Hove, East Sussex, UK: Psychology Press.

viii. Weinschenck, S. (2012). The True Cost of Multi-tasking. Psychology today. https://www.psychologytoday.com/us/blog/brain-wise/201209/the-true-cost-multi-tasking

ix. Kim, M. (2014). The effects of external cues on media habit and use: Push notification alerts and mobile application usage habits (Order No. 3634009). Available from ProQuest Dissertations & Theses Global. (1612601926). Retrieved from https://search.proquest.com/dissertations-theses/effects-external-cues-on-media-habit-use-push/docview/1612601926/se-2?accountid=14784

x. Wohllebe, A. (2020). Consumer Acceptance of App Push Notifications: Systematic Review on the Influence of Frequency. Kaposvár, Hungary: International Journal of Interactive Mobile Technologies. Retrieved from https://online-journals.org/index.php/i-jim/article/view/14563

xi. Pielot and Rello. (2017). Productive, Anxious, Lonely – 24 Hours Without Push Notifications. In proceedings of the 16th international conference on Human-computer interaction with mobile devices & services, MobileHCI '14. Association for Computing Machinery, New York, NY, USA. Retrieved from http://pielot.org/pubs/PielotRello2017-MHCI-DoNotDisturb.pdf

xii. Bowles, N. (2019, October 6). Addicted to Screens? That’s Really a You Problem. The New York Times. https://www.nytimes.com/2019/10/06/technology/phone-screen-addiction-tech-nir-eyal.html

Who Said What: A UX case study of empowering online shoppers