Wendy’s woes, Kate’s no-show, and cute Korean carebots
Wendy’s surges behind
You’ve probably heard of “surge pricing” and “dynamic pricing”. And you’ve definitely heard of “Wendy’s”.
A few weeks ago, those words all started appearing in news items when new Wendy’s CEO Kirk Tanner announced an investment in digital menu boards at US company-operated restaurants by the end of next year and added that the chain would be testing dynamic pricing. Big mistake! Cue outrage!
Mainstream media immediately misreported this as “surge pricing”, which is what rideshare app users encounter when it starts to rain and their Uber fare suddenly shoots up as demand increases. Although the difference is subtle, dynamic pricing is the raising and lowering of prices when there’s a limited supply of inventory such as airplane seats for a specific flight, or hotel room vacancies for a specific night.
What there is never a limited supply of in America is hamburgers.
You might find the real-world distinction between surge pricing and dynamic pricing debatable, and you wouldn’t be the only one. Customers were outraged and US Senator Elizabeth Warren called it “price gouging”.
CEO Tanner was left with a face as red as his corporate mascot’s hair and the Wendy’s Corporation had to clarify that prices would not be going up when demand was high but would decrease when store traffic was low. This might make some kind of economic sense. But where the company went wrong is that the strategy makes no psychological sense. Why? Because people hate negative unpredictability.
In fact, people crave predictability. A wide range of human behavior can be attributed to an instinctual avoidance of unpredictable situations. Predictability is an overarching factor in the development of neural networks, even in much simpler creatures than CEOs. Our brains are basically prediction organs that constantly model and remodel the world according to sensory input. Our emotions become unpleasant when that input fails to match the model. We feel stress, fear, and anger. We adjust our circumstances accordingly. We avoid the source of those emotions. None of this is what a brand wants.
The positive exception to unpredictability avoidance is what marketers call “surprise and delight”. Evolution has molded us to feel the positive emotion of excitement when we discover a new source of food in the forest or a deal at the store. (Unfortunately, this response also accounts for gambling addiction when we get hooked on the positive rush of dopamine from winning a turn.)
There are other contexts where we seek out unpredictability, such as thrill rides at amusement parks, but even then, their unpredictability is safely circumscribed and we laugh along with our companion about a sudden roller-coaster drop. Laughter is what we do to signal to ourselves and others that an unpredictable event did not turn out to be negative after all. Caveman version: “Lol, Grunk, that sudden movement in the tall grass wasn’t a tiger, it was a rabbit, and now you’ve pooped your loincloth!”
All this is to say that customers, who – unfortunately for CEOs and economists are human beings and not hyper-logical robots – react very badly to the idea that comfort food might become uncomfortably unpredictable where its price is concerned.
We are used to seeing dynamic pricing at online marketplaces such as Amazon. You might not be aware of it, but dynamic pricing even exists in supermarkets, with Canadian giant Loblaw rolling out the largest deployment of electronic sales tags (ESLs) in North America back in 2019. But potential gains in profits must be weighed against any impact on customer perceptions. Of course, this is where a brand can get into trouble.
Brand strategists often talk about trust. And I believe trust to be the number one factor in building brand equity. Ultimately, what trust comes down to is predictability. You trust someone when you can reliably predict their behavior in given situations. This is what makes corporations like Apple and Google risk-averse (and it’s what recently led Google to stop generating pictures of people in its AI image creation tool Gemini).
The flip side is that we tend to avoid people who behave unpredictably, and if we’re stuck with them we experience negative emotions. This is also true for brands. So in the lives of a whole bunch of fast food fans, that unpredictable person now goes by the name of Wendy.
Cute Korean carebots are coming
Rowan Cheung, founder of an excellent newsletter called The Rundown AI, has reported on an AI-augmented robot doll that South Korean local governments have supplied en masse to people suffering from dementia and seniors who live alone. The $1,8000 dolls are made by Hyodol and can hold full conversations, remind people to take medication, and send alerts to remote caregivers when no movement has been detected for a certain period. Around 7,000 dolls have been distributed by Korean authorities in a bid to handle elder care in an era of declining birth rates and an aging population.
Of course, it’s normal to have an initial reaction of, “Oh my god, this is so dystopian, I might as well be watching Black Mirror (yet again!), and what a shame that society has come to this, with care being outsourced to robots instead of helping people to engage in genuine human interactions.” However… if you have ever been with somebody who is suffering from dementia, you would know how exhausting it can be after a very short period of time. Family members may not always be available (or even exist) and in-person care is often unaffordable. Not only that, but initial tests seem to show that people who use the dolls show reduced depression levels and take their meds when they should more often. Yes, we are living in strange times, but extended life expectancy is a positive outcome of scientific advances, and, like everything, brings with it new challenges.
TikTok faces big block
US lawmakers are famously clueless about tech, starting in the early days of the internet and continuing into the social media era. They are now being asked to legislate on whether TikTok should be subject to a ban in their country, so obviously some of them must be scratching their heads. But in an unusual case of bipartisan agreement (maybe American politicians should scrap everything else and restart the country based on their united opposition to TikTok!) a bill that would force parent company ByteDance to cut ties with the app within six months or go dark was passed overwhelmingly by the House of Representatives yesterday. The bill eased through after various intelligence agency heads spoke at a House intelligence committee hearing on whether the CCP could use the immensely popular app to influence the upcoming 2024 US elections, with FBI Director Christopher Wray telling representatives, “Americans need to ask themselves whether they want to give the Chinese government the ability to control access to their data.”
As the bill arrives at the Senate, it’s unclear whether support will be quite as unanimous as in the House. Ohio senator J.D. Vance has said that he is “generally sympathetic to the idea [that] TikTok is sort of a [Chinese Communist Party] spy app,” while his colleague Cathy McMorris Rodgers from Washington commented, “It's important that the members have an opportunity to get this classified briefing from Department of Justice and other intelligence agencies around what they see as the threat of apps that are owned by foreign adversaries.” But do they know what the For You feed is?
Members of Congress are being lobbied hard to trash the initiative by TikTok users and influencers but this legislation would not be the first to target the app, which has been banned since 2020 in India and also faces legal hurdles in Canada, Europe, and elsewhere. Although the reaction in China appears to be supportive of ByteDance, it’s worth noting that China currently blocks US websites, platforms, and apps such as Gmail, Google, YouTube, Wikipedia, Facebook, WhatsApp, Instagram, and Twitter. Freedom of speech is a tough concept for totalitarian regimes. Interestingly, this political debate comes in the same week that tech bloggers uncovered hidden TikTok code that suggests the company is on the verge of releasing a photo-sharing app as a direct competitor to Instagram. Looks like Xi might have to challenge Zuck to a cage match, with winner takes all.
Airline chatbot grounded by judge
In an intro to this newsletter last September I wrote about how chatbots can hurt a brand. Now there’s more evidence that corporations need to take extra care when deploying automated agents on their platforms. In a judgment by a small claims court in British Columbia last month, Air Canada was found guilty of Negligent Misrepresentation when the chatbot on its website communicated incorrect information to a passenger that resulted in him misunderstanding the mechanism for obtaining a bereavement fare following a relative’s death. The judge wrote: “In effect, Air Canada suggests the chatbot is a separate legal entity that is responsible for its own actions. This is a remarkable submission. While a chatbot has an interactive component, it is still just a part of Air Canada’s website. It should be obvious to Air Canada that it is responsible for all the information on its website. It makes no difference whether the information comes from a static page or a chatbot.”
What the judge made crystal clear is that organizations cannot divest themselves of legal or moral responsibility for the output of the chatbots they use to communicate with people. And, as we know, chatbots hallucinate. Using them is therefore not only a brand safety issue, it is a legal hazard.
Jordan LeBel, food marketing professor
I’ve been a vegetarian since I was 19 years old. I don’t go around trying to convince people that animals are nice and we shouldn’t kill them just because we’re hungry and like the taste of them (I like the taste too!) but when Jordan LeBel, a leading expert on food marketing, agreed to be interviewed for Discomfort Zone, I jumped at the chance to ask him a question about meat substitutes. He graciously took time out of his busy schedule to think through this issue and share his opinion.
Q. Lab-grown meat is better for climate change and animal welfare than traditional livestock but might be a tough sell for carnivores. Would you have any insights on how to market these new food products to meat lovers?
A. In this case, much can be learned from the trajectory of the plant-based meat category. First, it was an epic branding failure on many levels, starting with the overt competitive positioning against meat. This instantiated mental associations and comparisons to a reference category (meat) that often resulted in barriers to trial and adoption. Secondly, a compelling “buy-me” value proposition should not convey “it tastes like” but rather “it’s an entirely new taste experience”. Thirdly, the scope of innovation and differentiation within the plant-based meat sub-category has been limited, with most options having a similar appearance and taste. To succeed, lab-grown meat needs to bring more differentiation to consumers’ dinner tables with tastier ready-to-cook or ready-to-heat convenience. Oscar Mayer might have figured that out: the company recently announced that it will introduce plant-based sausages and hotdogs, acknowledging that previous offerings in that category failed to deliver “largely due to disappointment in (…) taste and texture.”
Also, save-the-planet appeals, if used (and I am not convinced they need to be), are tricky. Framing the issue too widely makes consumers feel helpless, while framing it too narrowly reduces the magnitude of the problem. For instance, a claim that food waste costs the global economy over $1 trillion is impressive and conveys the magnitude of the issue but leaves consumers puzzled as to what their individual actions might have on the problem. Alternatively, emphasizing that food waste costs each Canadian household about $1800 per year makes the problem more relatable though it does not convey the size of the problem or the urgent need to tackle it. Proceed with caution.
On the packaging front, lab-grown meat needs to be more creative and avoid the “green=healthy” trap. In one notable fail, a manufacturer used turquoise on its faux-meat packaging. Why, oh why?! And since many purchase decisions are made at the point of selection, merchandising, including in-store sampling, should be deployed with renewed creativity to move consumers to trial. I think here sensory seduction should be the main goal. Overcoming the perception of a lack of naturalness is key.
Let’s circle back to the need for lab-grown meat. In her book Technically Food: Inside Silicon Valley’s Mission to Change What We Eat, Larissa Zimberoff paints a somber portrait of tech-focused solutions intended to move food production to the lab, arguing that such innovations are a barely veiled attempt to cut farmers out of the equation. Having grown up on a farm, I’m particularly sensitive to that argument. So perhaps what we need is not more lab-grown meat but to reconnect with farmers and those who produce our food and to learn to value their hard work.
In an unrelated category, I’ve been admiring the inspiring work of Equifruit, an importer and distributor of fair-trade organic bananas. To widen its appeal, the company decided to focus on the “mindful-ish” consumer, an acknowledgment that the targeted buyer doesn’t always behave and consume in a mindful manner, leading to a more inclusive messaging, even one using humor (unusual in the category). The recently updated visual identity is also as stunning as it is ground-breaking.
In the end, marketers have a Herculean task ahead to overcome Bennett’s Law, which states that with rising income, families and nations tend to replace cheap low-cost carbohydrate calories with more expensive foods, chiefly land-based animal protein. Countries with higher GDP are now eating a diet composed of 60, 70, or even 80% animal protein (as a percentage of total calories). At the risk of sounding too critical of my own profession: we need to be way more sophisticated in our approach if we are to move the needle on this one.
Jordan LeBel is a professor in the Department of Marketing of the John Molson School of Business (Concordia University) where he teaches the popular elective course “The Marketing of Food.” A former chef, Jordan taught food service management in Norway, Switzerland, and the United States at Cornell University’s famed School of Hotel Administration. Jordan is a member of the standing scientific committee on obesity prevention at Quebec’s National Institute of Public Health.
Seeing is disbelieving
In a perfect storm of celebrity gossip, conspiracy theorizing, and tech disruption, a photograph released by the British royal family that was intended to allay fears about the health of one of its members has backfired more spectacularly than Harry and Megan’s Netflix “production” deal.
As this article in Fast Company points out, the general public has become highly skeptical of photographic veracity after years of photoshopped models and months of AI deepfakes. In an era when image manipulation can be carried out by a 10-year-old with a smartphone, the perception of photography as a representation of reality appears to be waning.
In other monarchy news, I learned this week that James I of England (who was also James VI of Scotland in the same way that Elon Musk runs Tesla and SpaceX) had a succession of male lovers that he made no attempt to hide. The only reason I never learned about this in school is that the Victorians were homophobic prudes, so it’s a good thing they didn’t have Photoshop, although Victoria’s son George V did have to give the family a brand makeover when their German cousin turned out to be a warmonger.
When I fact-checked that last piece of info, I also discovered that King Charles and his sons’ surname is actually Mountbatten-Windsor, not Windsor. To complicate things even further in the branding realm, “Mountbatten” was also anglicized to hide its German origins during the First World War. From the battlefields to the tabloid pages, life is certainly a struggle when you’re a royal.
Spies vs AIs
Did you know that the CIA now has a Chief Technology Officer? It’s definitely not a state secret because the person in question, Nand Mulchandani, was interviewed for this week’s episode of the a16z podcast.
Are you wondering whether spy agencies are building their own AIs? As Mulchandani points out, “There’s no app store for spy software.” Does using AI to create deepfakes and convincingly impersonate people make the CIA’s job easier or harder? The spy boss has the answer. I found this a fascinating and revealing peak behind the scenes at the CIA. Or should that be the CAI?
Vicious cycle?
Things are sometimes overwhelming for Midjourney. Maybe when it produced these pictures the image generation AI hadn’t slept the night before or asked one of its kids to do the work for it. Either way, this output probably didn’t meet the user’s expectations.
Send me your comments on Discomfort Zone at john@johnbdutton.com and please connect with me on LinkedIn if you haven’t already.