Figure 1. Pizza. (“Pizza pie,” jeffreyw, CC BY 2.0, via Wikimedia Commons.)
A friend of mine had just returned home from a bad day at work as a healthcare provider. Not the kind of bad day we have in technology, but a first responder kind of bad day. The kind of day that requires a glass of wine, a movie, and a pizza to take her mind off things.
She went online and ordered a pie from her favorite pizza delivery place and turned on the television. The online order tracker reassuringly told her dinner would arrive soon. Five minutes left! She got out the plates and napkins.
At fifteen minutes past the delivery time, the order still had not arrived. The online delivery tracker said her order was in her building, but it had not been delivered. The delivery guy must have been held up in traffic or something. A half hour later, she tried calling the restaurant, but no one picked up. No one at the pizza shop? Something was very wrong. After two hours of trying to get through, she gave up. She ate cold leftovers and wondered if a major delivery chain had stolen her pizza money.
Going Up the Ladder
As surprised as I was to hear the story of the undelivered pizza, what happened next is what set off my UX spidey senses. After the shock had worn off, she tried calling the corporate headquarters to complain about this rogue local pizza place. This was a major pizza chain with a reputation for fast, reliable delivery. Surely such an outrageous service failure would warrant a direct line to the VP of customer service, she assured herself.
But she was stonewalled here as well. The website had presented her with endless choices of cross-linked pages leading her back to the central customer support number. Amazingly, that number was identical to the one for the local store she had tried calling. The website had recognized where she was calling from and had routed her back to square one: the local store. The online chat and email for customer service had led to a form letter that said, “someone would return her message” at some point in the future. She never did receive a response to any of those messages.
When she had finally managed to get through to the store, the day manager had offered her the only explanation she would ever get from the company, “Yeah, that order was cancelled,” and had hung up on her.
Figure 2. Labyrinth at Chartres Cathedral. (“Labyrinth at Chartres Cathedral,” Daderot, CC BY-SA 3.0, via Wikimedia Commons.)
Technology Moats: Defined
This incident led me to investigate a phenomenon I have come to call a “technology moat.” A technology moat is a subcategory of dark patterns when companies use technology-enabled customer service to hinder customers from contacting them in order to prevent complaints and returns and other types of financial compensation. Technology moats are slightly different from other dark patterns; they are not intended to trick customers into providing benefit to a company but are intended to prevent losses from a corporate balance sheet. The common denominator between dark patterns and technology moats is the use of UX design principles in a negative way to benefit the company and/or hinder the customer.
Unit Hassle Costs
Returning to the example above, what amazed me was the intentionality and completeness with which this company had obstructed my friend’s complaints. I currently work for a company where customer service is a deep core value; this idea of obstructing a customer seemed impossible, an anathema, to me. Corporate responsibility had been pushed down to the franchise level, apparently the company had decided that blocking this complaint—really all complaints—was more profitable than hiring customer support staff and processing refunds.
This incident had all the earmarks of a dark pattern. So, how might this qualify as a dark pattern? While technically the company could claim that customers are provided with multiple channels to reach them, in reality they are insulating themselves from their customers to the point where any meaningful contact is impossible, i.e., a technology moat.
While this concept may sound farfetched, it is actually a commonplace practice. Recent research by Anthony Dukes and Yi Zhu found that “many customer service organizations (CSOs) reflect a tiered, or multi-level, organizational structure, which we argue imposes hassle costs for dissatisfied customers seeking high levels of redress.” Dukes and Zhu developed a mathematical model to calculate “unit hassle cost” which they define as “the level of annoyance or frustration that an individual experiences should she be inconvenienced.” This “unit hassle cost” is a calculation of the financial cost to a company for introducing some level of friction into the customer support system. It allows the organization to decide how much hassle they can get away with before the call becomes unprofitable. (It is in many ways similar to the “hassle cost” Anja Lambrecht and Catherine Tucker discussed in their research on customer contract setup.)
As presented in an article from the University of Minnesota, Dukes and Zhu’s study suggested the following:
- “the more the hassle, the less likely a customer would escalate a less severe claim and the more likely it would mitigate illegitimate claims;
- additional hassles may help companies better control costs tied to customer complaints, such as reimbursements or repairs…”
Customer experience expert and Forbes contributor Shep Hyken in his article “Putting Hassle and Friction Into Customer Service” summed it up well: “In other words, creating a hassle is more profitable to a company than making it easy for customers to get the refund or compensation they may deserve.”
Benefits of Automation
Please do not get me wrong—I am not against automated customer support. I have been in software development a long time and know how essential automation is for both the customer and the producer. In fact, it is not much of a stretch to say today’s global economy would have a very difficult time functioning without it. In addition, many customers prefer the instant access of automated support of one kind or another, especially for less complex issues.
Companies as well have reaped the benefits of this automated, non-human support for years in the form of huge cost savings. In fact, some companies have the theoretical goal of trying to achieve zero customer calls or even removing human calls altogether. And the end result of this automation is a growing technology layer between the company and the customer: a buffer allowing customers to get answers quickly without having to wait in endless queues for help and allowing customer support representatives to focus on tasks that require human intervention.
Figure 3. Leeds Castle, Kent. (“Leeds Castle, Kent, England 3 – May 09,” Photo by David Iliff, CC BY-SA 3.0, via Wikimedia Commons.)
While technology buffers are necessary and beneficial to both the consumer and producer, there is a point where companies can use them defensively to hinder or prevent customers from receiving support. It is when this happens that the gap becomes a technology moat. Just as moats were used in medieval times to insulate the inhabitants of a castle from external threats, a technology buffer becomes a technology moat when they are used to prevent interactions that might prove expensive or embarrassing for a company.
A technology moat is more than just a poorly implemented customer service system. Many companies have poor customer support records due to lack of funds, training, or research. It is also not a matter of the extent of the automation. Amazon is an excellent example of a fully-automated customer support system that goes to great lengths to ensure their customers are treated fairly with the least amount of friction. Conversely, companies with human customer support representatives could make the experience so frustrating and time consuming that customers eventually give up.
So, how can you spot the difference between a beneficial automated support system and a technology moat? It has more to do with the ultimate goals of the customer service journey. Is the customer journey designed to make you as delighted as possible? Or are there intentional delays, friction, or annoyances built in to discourage or hinder the customer?
I am tempted to say that there is a clearly defined point where a line is crossed and you could pin a dark pattern label on a technology moat as you can with many other dark patterns. In fact, the EU passed a law prohibiting such practice in 2012. But given the discussion above on unit hassle cost, I suspect it is more of a spectrum, with companies deciding how much hassle they can afford. Our pizza delivery service is an example of a company on the negative end of that spectrum.
Lower the Drawbridge
I started with a detailed description of my friend’s story because I wanted to drive home the emotional impact of these practices on the customer: the human reality happening on the other end of the phone or keyboard. The lasting damage to the relationship between consumers and businesses is the price that unit hassle costs do not factor into the equation.
Of greater concern to me is the potential for abuse of automated customer support as our reliance on it increases. From the customer perspective, the current COVID-19 fueled environment is driving populations online in unprecedented numbers. More and more, they consider the internet an essential part of their day-to-day lives for basic necessities for everything from telehealth to groceries and education.
So, as our level of dependency for online services is increasing, our ability to be treated fairly by service providers is decreasing. As consumers, our means for recourse are restricted by the very technology that was created to help us.
A great deal of research and advertising dollars have gone into understanding and building service relationships. Companies that do understand this will go to great lengths to establish a long-term emotional connection with their customers. Those companies also understand the long-lasting damage negative experiences can have on their bottom lines, their brands, and their reputations. Nowadays, customers are not keeping any of this to themselves. A NICE CX Benchmarking report found that
- 83% of customers who had a positive experience share it on social media,
- 89% will buy more products from a company with a positive experience, and
- 81% said they would switch to another company after a bad experience.
The good news is that consumers will share positive experiences if businesses meet them half way. There are many excellent articles on the benefits of good customer support or the damage done to brands and reputations by bad customer support.
For UX professionals who are dedicated to reducing the friction between customer support and customers, I have this advice to share from a recent private conversation with Shep Hyken (Forbes) as part of this research:
- Start by mapping the customer journey.
- Identify those activities that could cause customer friction.
- Look at the drivers for each of these touchpoints and identify what is really happening behind the scenes.
- Identify those friction points both internally and externally. There might be friction points behind the scenes that are spilling out into your external experience (e.g., software or procedural steps).
- Take a look at your favorite sites for customer experience. Examine what it is that they do that your organization is not doing. It does not necessarily have to be about convenience, but something they are doing right that you can learn from.
While we might need to build a technological barrier to survive in this online environment, lowering the drawbridge of access will become an essential element in keeping your customers happy now and into the future.
Retrieved from https://uxpamagazine.org/technology-moats-the-dark-pattern-of-intentional-friction-barriers/
Comments are closed.