Launching 12 years ago in Sweden, Spotify has revolutionised the way people listen to music. Replacing the piracy of Limewire, Napster, and The Pirate Bay, Spotify heralded the beginning of the trend of subscription listening, while also becoming the go-to music streaming site. And according to the company’s Q1 2019 report, it rakes in 217 million active monthly listeners, with half of those users paying for a premium subscription. 
While Youtube, Soundcloud, and Apple Music offer similar services, a 2018 Vetro Analytics survey of loyalty and engagement revealed Spotify to be the site that people tend to stick with.  What’s driving this loyalty? In part, the Discover Weekly playlist – since July 2015, the playlist has been a mainstay of the listening experience, gifting its users a two-hour-long personalised playlist of new tracks each week.  Considering one Monday in September 2015, when a technical hiccup delayed its release, people took to Twitter distraught, it’s a product listeners have come to expect to be delivered. 
“When you’re designing a digital product, you want to create habit-forming products,” says Milena Bursztyn, MBA candidate at the University of Chicago Booth School of Business. “It’s also one of the hardest things to do and where behavioural science really comes in handy.”  So how does behavioural science help form people’s habits? As this year’s insights partner for Nudgestock 2019 – the UK’s largest festival of behavioural science – Canvas8 spoke to Bursztyn, who presented on the event stage with Will Page, Spotify’s chief economist, to find out how digital brands can use behavioural science to promote loyalty.
The fresh start effect
“One of the behavioural reasons [for the playlist’s popularity] is that it’s released on a Monday and, by the end of the week, it’s been refreshed. Monday is the day you start something new – it’s a mental exercise,” explains Bursztyn. “Think about when you start a diet – it’s never on a Saturday.”  Aspirational consumers can be described as people who ‘strive for more’ – an attitude common among Gen Zers and younger Yers, and a quality possessed by 37% of global consumers.  According to a survey by Deezer, aspirational behaviour is common among music consumers – almost half of the 1,000 Britons surveyed by the streaming site said they wished they had more time to discover new music. 
Using temporal landmarks – days that hold significance – in a product’s design can help people achieve aspirational goals. Behavioural scientists have coined this phenomenon ‘The Fresh Start Effect’, describing the psychological feeling of having a clean slate. Whether making a commitment to call your mum every Sunday or choosing to quit smoking in January, temporal landmarks help demarcate boundaries of time. In turn, this helps people to organise experiences and activities so they can better form habits.
With aspirational consumers 1.3 times more likely to have used Spotify in the last month, when it comes to digital loyalty, recognising these behaviours is important.  It’s a tack that's also been taken by Nationwide in their recent ‘Pay Day = Save Day’ campaign, encouraging the quarter (25%) of Britons who currently have no savings to think about their finances.  By flipping the narrative associated with Pay Day, Nationwide is aiming to help the 46% of people who regret not putting more money into their emergency fund to create positive saving habits. 
Inside Weather (2019) ©
Creating a perishable good
Digital scarcity helps drive loyalty by minimising people’s choices – rather than 50, or 100, new songs to listen to each week, Spotify delivers 30. “Think about a concert where your favourite band is performing for only one night – you have only one shot. In economic terms, you’re more likely to go and pay a high ticket price than if the band plays every weekend,” says Bursztyn.  Scarcity helps add value to a product by making items feel limited and desirable – it’s partly why streetwear drops receive such hype. And given that people can experience choice paralysis when faced with too many options, the playlists’ architecture helps people expand their musical horizons with minimal input and minimal decision-making.
Available for only one week, people’s options are further limited. “Listeners know they have one week to listen to the Discover Weekly playlist – the notion of scarcity is really strong here. It’s the fear of missing out – we don’t want to miss out on a good thing, whether those are digital experiences or in real life,” says Bursztyn. “A music playlist or product that’s available for a finite amount of time gives people a nudge to engage with it.''  According to Spotify, its Discover Weekly listeners stream more than double the amount of music than those who don’t check out their curated playlist. 
It’s no different in the workplace, either. “Slack is where work happens,” boasts the company’s website. “[It] brings people, data, and applications into a single place where people can effectively work together.” The workplace instant-messaging service – a mainstay of office life for more than 10 million people – similarly engages people through principles of scarcity.  “Slack is all about peer pressure and network effects. If your colleagues are on Slack and that’s where a lot of the work happens, then you’re just naturally pulled onto the platform,” says Bursztyn. “Again, that’s FOMO – imagine being the only person in the office who’s not on Slack. When the time comes for a promotion, that interaction and involvement on Slack would be considered.” 
We don’t want to miss out on a good thing, whether those are digital experiences or in real life. A music playlist or product that’s available for a finite amount of time gives people a nudge to engage with itMilena Bursztyn, MBA candidate at the University of Chicago Booth School of Business
Personalisation isn’t new, but it’s key to digital loyalty. “[The playlist] is unique because it’s something personalised people want to share with their friends. You become loyal to digital products and you want to support the business and you want to engage with the product,” says Bursztyn.  Almost half (49%) of people expect specialised, bespoke treatment for being brand-loyal, and 36% would be willing to pay for personalised products or services. But there’s a gap between what consumers expect and how companies deliver on this – only 7% of organisations listed personalisation as their top priority in Econsultancy’s 2018 Digital Trends report. 
Designing digital products with the users’ environment in mind can help to generate more nuanced personalisation. In March 2019, Spotify introduced ‘Happy Hits’ and ‘Songs to Sing in the Car’, a careful consideration of the psychological and physical contexts where music is listened to.  Machine learning and AI can deliver that nuanced personalisation. Fintech firm Upstart uses machine learning alongside non-traditional variables – like job history and education level – to measure a person’s financial idiosyncrasies, potentially proving more personal than traditional human-led system
“The user-generated nudges are created in a feedback loop – people play songs, follow playlists and other listeners’ favourite music, and then Big Data is collected which informs Discover Weekly,” says Bursztyn.  And with more than 200 million Spotify subscribers to cater to, data-generation is the driving force behind delivering personalised playlists en masse. But it’s the human touch that adds finesse. “Having a little bit of familiarity is key to building trust. It can be exhausting to just listen to stuff you've never heard of before," said then-Spotify Discover Weekly manager Matt Ogle back in 2015.  So it’s the blend of known and unknown artists that helps join the dots between personal taste and making valuable listening suggestions.
jose aljovin (2019) ©
“Visual cues are really big in the digital product space. We all know people have very finite attention spans so when it comes to designing a product, you want to leverage all kinds of cues – and the visual ones are often the most effective,” says Bursztyn. Far from being a blind approach, combining data and behavioural theory can prove effective. “When people go to their Daily Mix playlist, they see a mosaic of artist images, signalling what the music contains. But our data was telling us that those collages weren’t doing the job. Instead, the insights revealed that we might be better off moving to a single image. It sounds pretty simple, right? But that one image had a huge impact on the user experience. When there was an image of a familiar artist, the user metric just skyrocketed.” 
As with personalisation, familiar visuals can drive loyalty and engagement, too. With 92% of global internet users using emojis in their daily communication, brands are harnessing this familiarity, using emoji marketing to boost brand loyalty.  Digital banking service Monzo integrates emojis into different aspects of its service, from user reactions to service notifications. While this communication may seem counterintuitive for a banking app, Monzo’s position as a challenger brand – one that puts an emphasis on the consumer’s needs – positions it well in the space of familiarity. Similarly, Domino’s benefited from implementing emoticons in its pizza delivery service via Twitter because the emoji worked as a natural extension of the brand’s warm vibe. And research from Return Path revealed that on Valentine’s Day, use of the lips emoji boosted email marketing read rates by four percentage points, compared to those that used text alone (24% and 20%, respectively). 
“If a product has visual cues that are able to communicate maximum information with the smallest of visual cues, they’re onto something very big,” says Bursztyn.  But visual cues can backfire, leaving a sour taste in consumers’ mouths. In 2018, Netflix’s ‘artwork personalisation’ was accused of inaccurate ethnic representation, following complaints from black viewers when they saw film artwork showcasing black actors who play minor roles, rather than leads.  Given that research by Netflix found artwork to be the biggest influencing factor on selection, artwork personalisation is a logical move.  But misrepresentation is problematic since it makes people feel like Netflix is shortcutting representation via tokenism. “It does happen when you have algorithms. We think of computers and algorithms as perfect machines, but at the end of the day, they’re designed by human beings,” says Bursztyn.  Encouraging diversity in AI’s creation, however, can help mitigate these biases.
Nate Johnston (2019) ©
Insights and opportunities
Blends of behavioural science, big data, and consumer insight drive loyalty in the digital sphere. By mixing scarcity with smart design, brands can take advantage of the powerful effects of behavioural concepts, which may have positive effects on user engagement. “You want to leverage big data, behavioural theory, and consumer insights to design these strategies,” says Bursztyn.  Spotify leads with the three R’s – regularity, randomness, and reinforcing – to promote loyalty, which creates regular temporal landmarks for release dates, injecting some randomness to keep products fresh and exciting, which serves to reinforce user interest.
However, when using behavioural notions of scarcity and FOMO, companies should err on the side of caution. With over two-thirds (68%) of American internet users admitting to feeling pangs of social media envy, feelings of FOMO can have a negative impact on people’s lives.  Especially since they can lead people to think they have a lower social status, which in turn creates feelings of inferiority.  To mitigate this, ensuring that behavioural nudges are used ‘for good’ can help to foster positive results.  “Richard Thaler often talks about this, it’s part of his message that when you’re designing nudges, they should always be for good, because you never want to mislead or mistrust customers or users,” says Bursztyn. 
People use their digital spaces in both a personal and social way. While clicking ‘private session’ on Spotify can give listeners some privacy, listening becomes social when data is collected. User-generated personalisation, for example, helps to socialise these spaces. But people are still conflicted about how comfortable they feel sharing their data – 51% of people across 10 countries would be happy to exchange their data with companies if there were clear benefits.  Tapping into these attitudes are Qwant and DuckDuckGo, search platforms that prioritise privacy, as well as the kill switch on Google smart speakers. So while nuanced personalisation can help drive loyalty by meeting expectations, companies can better cater to their users by understanding people’s attitudes towards privacy and how they want their data to be used.
‘Spotify Usage and Revenue Statistics’
‘Apple Music, Not Spotify, Ranks as the Most Popular Music Streaming Service’
‘Spotify’s new feature will help you fight the Monday blues’Fortune
‘Spotify’s Discover Weekly updated late and people were furious’
‘Aspirational Consumers: Why they Matter and What Brands Should Know’
‘We stop discovering new music at age 30, a new survey suggests — here are the scientific reasons why this could be’
‘A quarter of British adults have no savings, study reveals’The Independent
‘Three-quarters of Britons worried about financial situation, survey claims’The Independent
‘Spotify Opens Discover Weekly Playlist To Brand Sponsorship’
‘With 10+ million daily active users, Slack is where more work happens every day, all over the world’
’What Do Customers Want?’
‘12 stats that prove why personalisation is so important’
‘Spotify expands personalization to its programmed playlists’
‘Spotify is getting unbelievably good at picking music — here’s an inside look at how’
‘The Science of Emoji Marketing’
‘Emojis May Increase Read Rate; Won’t Automatically Land Your Email in the Spam Folder’
’Netflix Viewers Are Angered By Ethnic Misrepresentation’
‘Film fans see red over Netflix ‘targeted’ posters for black viewers’The Guardian
‘Jealous of your Facebook friends? You’re not alone’Experience Magazine
‘The Stress of Social Comparison’
How Can Behavioural Economics Nudge For Good?’
‘Consumers willing to share data in exchange for benefits, survey shows’