Accessibility in E-tail: The Effects of AI on CX & UI

Imagine a shopping frenzy on par with the chaos of Black Friday, but instead of a yearly event with major sales, pricey new apparel is released weekly. The most dedicated shoppers form lines outside of stores the day before a release, regardless of the weather. However, with the affordances of the internet and shifts in purchasing habits due to the global pandemic, an increasing number of these zealous shoppers have moved online, poised to make a purchase the moment of a restock. These individuals are not just seeking any products – such fanatics, known as “hypebeasts,” take lengthy measures to purchase specific apparel and footwear released by “hype” streetwear brands that are notorious for their limited releases and short inventory.

 

What is a hypebeast?

A hypebeast is often a young male who is passionate about sportswear and sneakers, similar to individuals known as “sneakerheads” who collect and trade rare sneakers. Usually motivated by celebrity influencers and hip-hop music, these individuals purchase and wear apparel from trending streetwear and sports brands, and will do whatever it takes to own the latest trendy item. While many popular “hype” brands originate from North America and the United Kingdom, the hypebeast community is globally located, with each geographical cluster consisting of local preferences and trends.

 

What is a bot & why are people using them?

As communication and marketing efforts increasingly move online, hypebeasts with faster internet access and deft hands are the ones who make successful purchases. When a few savvy programmers caught on to this shift, they began designing bots to navigate online purchases faster than humans. These bots are operated by machine-learning algorithms to assist the user in checking out faster than other shoppers. These bot makers have made their services available to zealous hypebeasts for a price, meaning that some individuals may choose to pay up to ~$200 for just a chance to purchase an item from a “hype” brand. 

 

Bots aren’t a completely new concept - concert ticket scalpers, bid snipers on eBay, and high frequency traders on Wall Street are all tangential agents that rely on fast internet, timeliness, and maximum purchase potential to complete a task successfully. Recurring issues of unfairness arise in each scenario where a human is competing with algorithms within a fixed amount of time.

 

The effects of bots on customers & retail brands

Bots have made it almost impossible for the average customer to compete and make a successful purchase in time. Not only are they extraneous to earnest customers, but the brands also lose profit as bot makers extract an intermediary rent to complete transactions as quickly as possible. Bots have complicated traditional methods of online shopping for hypebeasts, causing problems and raising issues of unfairness in the community. They have also affected the way retail websites are designed on both the front and backends, as brands continue to build their online stores in a way that defends them from bots, or at least slows down their efficacy.

 

EU sell out times - items sold out in 10 seconds

 

Artificial intelligence in online markets

Much existing scholarship about the implementation of artificial intelligence in online shopping lauds the ways that it facilitates the shopping experience (Serenko, Sharma, Sheth). However, the parasitic spread of bots in the hypebeast community deviates from this utopian view of computational assistance. Algorithmic disturbance raises issues of unfairness when humans can no longer compete with the instantaneity of machine work. While online markets preclude the buyer from having to be at the physical site of an auction, they also allow non-human actors to compete in the auction process. An ethical gray zone emerges when artificial agents can compete with humans in auctions. Those with access to such agents are at an advantage, while those who are uninformed or cannot acquire them continue to make failed purchases. 

One reddit user pleads “I failed on getting the yeezy 350 moonrocks, I stayed up all night, some people got 6 pairs and I failed at just getting one pair… please I honestly want to get one pair, ive never been successful with yeezys, Please help me out Im begging you” (Reddit).

 

Large-scale effects of bot usage

With more and more people using bots, the community has shifted on a larger scale. The aggregate use of bots exacerbates the difficulty of purchases and raises the total amount that an individual must spend on a given item. In a YouTube video that introduces sneaker bots, the user maintains that “bots have become essential to copping limited sneakers online. Although, even with the bot you might not be able to cop ‘em. Nowadays everybody’s using bots… Without a bot, you almost have no chance.” (Hypebeast TV). 

 

The emergence of these bots has created a hierarchy in the community based on money, since more advanced bots are available for higher prices. While the actual processes of these bots are undisclosed to the public, in general they work like an advanced autofill. The bot user simply instructs the program of their desired item(s) and size, and enters the relevant card information and address beforehand. The moment the online store opens, the bot performs the necessary actions and checks out within seconds. 

 

Effects on the retail brands & their websites

While using a bot indeed gives the buyer a better chance at a successful purchase, the aggregate use of them has shifted the ways that some brands sell their merchandise. Brands often block purchases that are deemed too fast for human performance, but sometimes these are made by skilled purchasers without the help of a bot, and their diligent efforts are rejected because of the suspicion created by the prevalent usage of bots. 

 

Rather than creating bots to truly assist the buyer, bot makers’ intentions are to extract profit from zealous hypebeasts who are willing to pay extra. A bot called “Supreme Saint'' was created by two young men, Matt and Chris, who charged $100 for their services. The ecommerce bot received 200 orders for a particular pair of Nike sneakers, making the duo an easy $20,000 in five seconds (Schwartzberg). Bots don’t need to match customers to products – hypebeasts already know exactly what they want to purchase. As such, the bot makers are a parasitic third party that rides on a predetermined transaction in order to make a personal profit. 

Supercop for IOS - https://supercopbot.com/

 

Legality of bots

The legality of bot usage is quite gray. California and New York have laws against bots that lobby event tickets, and hype brands can very well win a case if they sue a bot maker. However, bot makers remain anonymous, and most hypebeast brands do not want to “puncture [their] cool-guy image” (Schwartzberg). To combat the mass usage of bots, hype brands have been creating different ways to sell merchandise online. While bots are indeed “unfair,” it is not surprising that people made them – they are a way to get ahead. The relationship between market design and artificial agent design is closely intertwined, and to bid-snipers and bot makers, online sites have certain architectures that allow them to be rigged by skilled coders. Retail brands must consider how to make their goods accessible and purchasable to a diverse range of customers, not just those who can afford the extra “entry fee.”

 

Online markets have facilitated shopping for many individuals, as all that is needed to make a purchase is an internet connection and valid payment information. The benefits that information technology offers to traditional auctions are vast, but sometimes they can disadvantage both customers and retail brands. The disruption bots have caused for hypebeasts raises questions of whether online purchases should be automated or not. A hierarchy is created, and while it may benefit those at the top, “the point is not whether some people benefit. It’s that so many suffer. These models, powered by algorithms, slam doors in the face of millions of people, often for the flimsiest of reasons… They’re unfair.” (O’Neil 44). Sentiments of unfair play demonstrate a larger trend of issues that arise when humans and machines compete. 

 

Computation is often seen as a source of innovation with democratic means and minimal errors, but quite the opposite is shown in the consequences of bot usage in the hypebeast community. Difficulties in combating bots persist, as their creators and exact functions remain anonymous. With online commerce operating too quickly for human reaction, it is unclear what role the unarmed human will have in an environment teeming with bot users, but it is clear that bots will continue to persist.

Retail brands must continue to stay wary of new bots and come up with creative ways to sell their merchandise in order to prevent discouragement from earnest shoppers. In recent years, random raffles have become more common and are one way brands have tried to bypass the ever-evolving development of bots.

It’s important to understand customer frustrations and experiences in retail overall and in certain niche markets. Ongoing user research plays an integral and necessary part in keeping pace with ever-evolving technology. Speak to a Key Lime Interactive research expert about how we can help you learn from your customers and prospects.

 

 

Works Cited:

  • Berry, David. “Against the Computational Creep.” Stunlaw.blogspot.com, STUNLAW, 27 Mar. 2017, stunlaw.blogspot.com/2017/03/against-computational-creep.html.
  • Cypryjański, Jacek and Aleksandra Grzesiuk. "The Role of Signals in Online Auction Purchase Decisions." Folia Oeconomica Stetinensia, vol. 15, no. 1, Jan. 2015, pp. 53-68. EBSCOhost, doi:10.1515/foli-2015-0019.
  • Danforth, Chris. “Sneaker Bots: How Do They Work?” Highsnobiety, Highsnobiety, 28 Apr. 2016.
  • Karppi, Tero and Kate Crawford. "Social Media, Financial Algorithms and the Hack Crash." Theory, Culture & Society, vol. 33, no. 1, Jan 01, 2016, pp. 73-92. 
  • Leach, Alec. “The Most Valuable Sneakers of 2016 Q2.” Highsnobiety, Highsnobiety, 20 Jun. 2016.
  • Lizé, Wenceslas. "Artistic Work Intermediaries as Value Producers. Agents, Managers, Tourneurs and the Acquisition of Symbolic Capital in Popular Music." Poetics, vol. 59, 01 Dec. 2016, pp. 35-49.
  • Marcoux, Alexei M. "Snipers, Stalkers, and Nibblers: Online Auction Business Ethics." Journal of Business Ethics, vol. 46, no. 2, Aug2003 Part 2, pp. 163-173.
  • Morris, Jeremy Wade. "Curation by Code: Infomediaries and the Data Mining of Taste." European Journal of Cultural Studies, vol. 18, no. 4-5, n.d., pp. 446-463.
  • Ockenfels, A. and A.E. Roth. "The Timing of Bids in Internet Auctions: Market Design, Bidder Behavior, and Artificial Agents." AI Magazine, vol. 23, no. 3, 01 Sept. 2002, p. 79-87. 
  • O'Brien, Matthew. “Everything You Need to Know About High-Frequency Trading.” The Atlantic, Atlantic Media Company, 11 Apr. 2014.
  • O’Neil, Cathy. 2016. “What is a model?” Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy. (Crown), pp. 26-44.
  • Sajonas, Felson. “Supreme x Louis Vuitton Box Logo Hoodies Have Resell Prices of up to $25,000 USD.” HYPEBEAST, 1 July 2017.
  • Sandvig, Christian. “Auditing Algorithms: Research Methods for Detecting Discrimination on Internet Platforms.” ICA Preconference, 2014.
  • Schwartzberg, Lauren. "American Hustlers." Wired, vol. 25, no. 6, 06, 2017, pp. 86, ProQuest Central; Research Library
  • Serenko, Alexander, and James, Hayes. "Investigating the Functionality and Performance of Online Shopping Bots for Electronic Commerce: A Follow-Up Study." International Journal of Electronic Business, vol. 8, no. 1, Jan. 2010, pp. 1-15.
  • Sharma, Amit. “How Predictive AI Will Change Shopping.” Harvard Business Review, 18 Nov. 2016, hbr.org/2016/11/how-predictive-ai-will-change-shopping.
  • Sheth, Beerud. "9 Ways Shopper Bots Will Disrupt E-Commerce." Marketing News, vol. 51, no. 9, Oct. 2017, pp. 3-4. 
  • Torres, de N. "Brands in High Demand should Sell Exclusively Online." University Wire, Apr 22, 2016, Education Database; ProQuest Central.
  • Trevathan, Jarrod, and Wayne, Read. "Disarming the Bid Sniper." Journal of Electronic Commerce Research, vol. 12, no. 3, Aug. 2011, pp. 176-186. 
  • Hypebeast TV. “Sneaker Bot Part 1 - What Are They & Why You Need One.” Online video clip. YouTube. N.p., 16 January 2016. Web. 15 October 2017.
  • Mandrew. “How To Cop Any Hypebeast Sneakers Manual (Without A bot).” Online video clip. YouTube. N.p., 17 August 2017. Web. 15 October 2017. 
  • Supreme Leaks News. “EU Sell-Out Times.” Instagram, 19 October 2017.
  • Reddit. “Lets talk about bots.” Reddit, 2016.
More by this Author
Christine Sim

Christine is a user experience researcher with a passion for product-led growth and a curiosity for understanding what makes people tick. She has a strong understanding of the strengths and shortcomings of qualitative and quantitative research methods, including when and how to apply them during the design process. Her appetite for interdisciplinary collaboration and experience working on fast-paced product teams has honed her ability to efficiently tackle complex problems. Christine’s industry experience includes: Financial services, insurance, risk management, eCommerce, DTC meal services, and technology, with a focus on B2B SaaS products. She has a unique talent for communicating and contributing ideas across all levels of stakeholders, using her avid storytelling skills to influence product and design decisions in a memorable and captivating way. Recognized for her poignant ability to get to the core of user problems, Christine is a tireless advocate for users, user-centered processes, and the client. Christine has an MA in Sociology from Columbia University and a BS in Media, Culture, & Communication from New York University. She has a publication in the Princeton Journal of East Asian Studies and has written extensively about her fascination with the way technology continues to shape the ways we socialize, exchange goods & services, and make meaning of the world. She has fluency in Korean and elementary Japanese and Mandarin language skills.

Comments

Add Comment