Here’s mud in your A.I.

If you’ve been paying attention to the industry news, there’s been a LOT of chatter about AI, (artificial intelligence,) and its various applications. And some of them are really intriguing and useful. With the ability to run predictive diagnostics, artificial intelligence (better described as data-driven machine learning,) is ideal for applications that can benefit from robust and speedy automation.

As is our way, it doesn’t take long for something useful and intriguing to be repurposed into something base and silly. Case in point: AI is also being used now for some fairly sophisticated parlour tricks, like recreating the Mona Lisa.

(Moves soapbox to the foreground.)

But the application of AI in marketing, and most specifically in the creative process, is really (in this sentient blogger’s opinion,) an overreach.

There’s no data set for creativity. In fact, there are no rules. What makes something “creative” is that it is indeed CREATED. By a human being. Part of why we buy paintings, and music and novels and sculpture is because we know there’s a backstory of someone who sweat it out in a studio or at the typewriter. Someone whose fingers bled. Someone who made mistakes, and tried variations, and threw whole passages in a trash can. We celebrate that humanity and that pain and the entire process when we consume anything creative. Not just the end result.

The same is true in marketing. Writing anything – an ad, a blog post, a commercial script – is hard. It’s taking business rules and mandatories into consideration and asking a creative person to then do intellectual gymnastics, linking sometimes disparate ideas in unexpected ways, without a net. Can you write an algorithm for that? Sure, but the results are likely to be shit – the kind of shit a hack would conjure.

For instance, you’ve probably seen these kinds of promotions for having AI write your next blog post.

What’s really happening there? I don’t claim to have knowledge of any of their algorithms, but I’d bet my last dollar that these are search bots that crawl the web for every piece of content written in the last 10 years about a particular subject. They ingest this huge data set, pick out bits and pieces using sorting criteria that prioritize those with the most clicks, comments and instances, and then rearrange and reconstruct a new version for you.

That’s not creating. That’s recreating at best. And theft at worst.

ChatGPT, the latest and supposedly greatest iteration of AI language prowess, has added a sexy wrinkle into their algorithm. Instead of just swiping content and repurposing it, it adds a level of dialogue formatting to make it sound more conversational, and thus more natural and believable. It claims to “answer follow-up questions” and “admits its mistakes.” And given that it is machine-based learning, it gets “better” the more often it’s used. (It’s driving educators crazy, as students who use it can be deemed to be cheating, not researching.)

What a paradox: a machine learning model that improves the more often it’s used, but also degrades the craft proportionately in the process. [Some disclosures: OpenAI, the company behind ChatGPT is funded in part by Sam Altman, Elon Musk, and just got a billion dollar injection of capital from Microsoft.]

In a recent viral kerfuffle on social media, recording artist Nick Cave (throaty lead singer with The Bad Seeds) reacted to a song that was “written” by ChatGPT “in the style of Nick Cave.” He could not have put it any better:

“Songs arise out of suffering, by which I mean they are predicated upon the complex, internal human struggle of creation and, well, as far as I know, algorithms don’t feel. Data doesn’t suffer. ChatGPT has no inner being, it has been nowhere, it has endured nothing, it has not had the audacity to reach beyond its limitations, and hence it doesn’t have the capacity for a shared transcendent experience, as it has no limitations from which to transcend. ChatGPT’s melancholy role is that it is destined to imitate and can never have an authentic human experience, no matter how devalued and inconsequential the human experience may in time become.”

Okay, back to marketing, and particularly the art of writing good advertising. Part of the craft is developing a sense of counterintuitive thought and embracing lateral thinking. Another big part is the symbiotic relationship of the headline and the image. But so much of modern advertising is about contextualizing the brand into the cultural moment from which it arises.

Think of how many great headlines and taglines would never – could never – be written by AI, by virtue of this necessity for divergence and a sense of the cultural periphery.

“Think Small” from DDB was a radical notion at the dawn of the 1960s. As Americans – less than two decades removed from World War II – were upwardly mobile, one of the most pervasive (and comical) trends in automobile-dom was an oblique obsession with size. Bigger was better, and American cars looked like ocean liners. The nation was also wildly nationalistic at the time, and DDB’s assignment was to sell a small, quirky German car to this audience.

To get to the heart of these cultural undertones and suggest an opposite notion was a radical idea. And paired with the spare art direction (bless you Helmut Krone) and that sea of white space, it became a touchstone for our industry. Could AI ever reach that level of insight? Of rebelliousness? Of sheer chutzpah?

“Think Different.” When this campaign (from TBWA/Chiat/Day) was released in the 1990s, it was paired with images of pioneering artists, thinkers and doers, like Einstein, Picasso and Miles Davis. It also featured zero body copy. What’s so interesting about the head/tagline is that it’s not even good English. But it was a fine encapsulation of the Apple brand in that moment and what it stood for. Would AI dare to break grammar rules to create an emotional response?

I suppose it’s mildly ironic, or at least cheeky, that I chose two classic advertising examples that use the word “think” in the headlines. And perhaps that’s exactly what I’m driving at, and what Nick Cave was fuming at.

Maybe I’m an old fart, but I prefer the consternation. The suffering. The pacing and the waiting and the wondering if this “idea” has legs. Give me time to think about what’s going on in the world. Give me the sudden burst of insight when the neurons start to fire. Give me the end of the sentence that starts with “wouldn’t it be cool if…” Give me a person, thinking about another person, and getting them to actually think different.

There’s nothing artificial about that.

The reports of the cookie’s death have been grossly exaggerated.

“The death of cookies,” to borrow a phrase, is (mostly) fake news. I wouldn’t be concerned with this topic, except that so many reputable media outlets have been publishing misleading and sensational headlines about it.

Take a look:

Adweek sent an email special report with the headline:  “The Death of Cookies.”

AdAge with Cheetah Digital posted on social with the same frenzied phrase: “The death of cookies”

And Segment (Twilio’s customer data platform) went even further with this social gem: “Digital Advertising in a Cookieless World.”

But here’s the problem: It.  Isn’t. True.

So let’s get to the bottom of why all these authorities are spewing all this gunk.

The first thing that’s important to know is that there are two primary kinds of cookies:  these are called “first-party cookies” and “third-party cookies.”  (I know, it’s weird that there’s no “second party cookies.” Somebody got ripped off in this deal.)

First-party cookies are data files that are shared between your browser and any website you visit.  When you visit a site, especially a site that you may go back to multiple times, say a news site like NYTimes.com, a small text file is generated and stored on your browser. It contains information about you (nothing too personal, unless you choose to save passwords via your browser, and please don’t do that,) like the browser you’re using, the operating system, where you’re located (via your IP address,) and even the browsing history.  They were originally created to optimize the performance of websites.  Since there’s a history encoded in the cookie, the website does not have to fully reload each time you visit it – it sort of restores the previous session, and then updates the site with its latest content. This is why your shopping cart on e-commerce sites is “remembered” and preferences on other sites are stored.  Each time you go back to the site, the cookie is updated with more information.  So it’s a bit of a history log between your browser and a specific domain. 

Disclaimer: This is an oversimplification of first-party cookies, but it will serve to help distinguish it from the other type.

Third-party cookies are different in many ways. First, and most important, they are stored under a different domain than the one you are visiting. They are typically “shared” from the domain you’re visiting with – you guessed it – third parties. The most basic example is this:  you go and visit NYTimes.com to read the news, and you see banner ads on the top and along the right side from a brand like Toyota. Since NYTimes is a publisher, and has sold advertising space to Toyota, they may have also agreed to share your data, and allow Toyota to drop a third-party cookie to track your browsing behavior.  The thinking with this was “it would help Toyota to know what other kinds of sites readers of the New York Times might visit, and what their browsing behavior is, so we can build a better profile of potential targets.” 

Disclaimer:  this is also an oversimplification of third-party cookies, but it should serve to help distinguish it from the other type.

Okay.

So, the headlines you’ve been reading are misleading, because they leave out a very important qualifier: the only thing that’s “dying” is the third-party cookie.

Pushing out headlines like “the death of cookies” or “a cookieless world” is like saying “music is dead” just because we’ve banned Justin Bieber. It’s sensationalizing the story at best. It’s clickbait at worst.

First-party cookies are here to stay, and there’s no way they’re going away, since it would cut off hundreds of billions of dollars in revenue being transacted every year. First-party cookies form the basis of ad targeting, retargeting, (yes, you can still be retargeted via first-party cookies,) display advertising (banner ads,) most social media platform algorithms, and the mighty Zeus of them all, search engine marketing and its associated retargeting.

[The post ends there, but I have a few words to say about WHY third-party cookies are getting the axe.]

The whole kerfuffle over third-party cookies is generally about privacy, and not having one’s data shared without one’s knowledge. But who are we kidding? Our data is getting shared every day, all over the place, whether we a.) like it or not and b.) know it or not.  Did you ever Google yourself? I’m sure you didn’t actively and purposefully put all that information there – it was aggregated from across the web without your knowledge or consent. How do you think data marketing companies make money? They go and mine data they already have, or it’s getting shared with them from retailers, credit card companies, and others. Yes, GDPR legislation has been adopted, but all it effectively does is add another annoying click before I get to the content I want on any new website. Ugh. And if the last year has showed us anything, (since third-party cookies have started phasing out and Apple’s iOS tracking regulations have been adopted,) it’s that you can still run effective and even mobile-friendly ads without third-party cookies and nobody is any worse for wear.

This author’s preference is to have his data shared (no social security numbers, credit card numbers or bank account numbers, thank you very much,) so that my general Internet experience is more carefully curated and more fully tailored to my preferences. Killing third-party cookies was accelerated as a knee-jerk reaction to the 2020 election and fears of the “echo chamber” effect, and more sensitive issues like child welfare, gender identity, and other possibly incriminating privacy gaffes. All the while, forgetting that you can still be retargeted via first-party cookies. They could have worked that stuff out and still made the Internet an interesting and contextualized place.

If Toyota wants to follow me around for two weeks to find out that I’m not a fit for their brand, so be it.  At least I won’t see ads for the 2022 Camry hybrid anymore.

Facebook’s Meta transition. A mashup that proves hardware is the new tech.

Late last month, noted CEO Mark Zuckerberg announced that Facebook is changing its name to Meta, and changing its official stock ticker from FB to MVRS.  The name Meta is a shorthand for the metaverse, which is itself shorthand for an almost fully immersed online world, where people can play, work, and gather in groups in the virtual sense. Zuckerberg is betting big on building it, even though it’s been tried before. (More on that in a bit.)

That this massive shift away from one of the world’s most recognizable brand names comes amid a slew of scandals is indeed curious.  But let’s leave all the political soundbites and sexy headlines aside for the moment.  This is not about the Facebook Papers, nor about Russian disinformation, nor about Cambridge Analytica, or data collection, or facial recognition…man, they do have a lot of shit swirling around the campus out there, don’t they?

Nah, this smells like a big bet hardware play, plain and simple.

This whole Meta rename is nothing more than a cosmetic corporate restructuring that will now control Facebook and its other well-known brands, including Instagram, WhatsApp, Messenger and Oculus.  A lot like when Google changed their name to Alphabet, and rolled up all their brands, including Google itself, under the holding company.  (PS – only investors care about this stuff, and THEY still call it Google. And the stock ticker for the company known as Alphabet is…GOOGL.)

So why isn’t Zuckerberg saying that?

I have an idea. Maybe it’s because the metaverse isn’t a great idea.  Or, rather, maybe it isn’t a great idea to shelter it under the enormous loads of cash that the artist formerly known as Facebook has at its disposal.  It’s been widely reported that the year one budget is over $10 billion, and that 10,000 people, mostly in hardware, will be recruited to make it go.

When any entrepreneur wants to launch a new idea, especially a broad and ambitious one like the metaverse Zuckerberg envisions, it’s good practice to prove it can actually accomplish something on its own merits.  It’s a good practice to seek capital from investors and show milestones that prove the concept.  In the absence of that kind of oversight and objective grownups in the room that business incubator model provides, it’s just another lavish vanity project.  The Metaverse is to Zuckerberg what space is to Bezos, Musk and Branson: a vast unknown that he hopes to monetize.

And let’s remember two important things about Meta’s metaverse:
First, the road to the metaverse was paved by Second Life way back in 2003, a full year before Mark Zuckerberg’s “hot or not” turned into “thefacebook.” It is a metaverse full of avatars and provides an almost identical experience to what Zuckerberg envisions: an interesting alternative online environment, where you can have virtual meetings and other whatnots.  (Kinda mostly trying to ply a virtual shopping mall, though.)

Second, and far more interesting: Meta’s virtual world will require, not suggest, that you purchase some very real and very significant pieces of hardware to access it. Oculus VR goggles are currently retailing at around $300, and may not have the full range of capabilities to access what will eventually become the Meta metaverse. It’s a long way to go to sell a bunch of accessories, but it sure sounds like a hot hardware play, doesn’t it? Build the metaverse, get a lot of good press, then tell those who can afford it that the only way to get on board is to buy some rechargeable VR binoculars, now available in six avatar-worthy colors!  All of this is coming right on the heels of Facebook (can we still call it that?) inking a deal with Ray-Ban to sell some fancy Smart Wayfarers that take photographs and play tunes, also for about $300.

If I didn’t know better, I’d swear Zuckerberg was trying to emulate Steve Jobs in some way. After all, Apple’s most successful product was/is the iPhone, not the Macintosh, its former flagship. It required the purchase of a significant piece of hardware. It was an ambitious project and came decades after the company launched. And Jobs didn’t just have the phone developed with a base OS and software.  He outsourced the smartphone “experience” to third party developers via the app ecosystem so every user could customize their device to their liking and have a uniquely personal interaction with it. It’s what ignited the phone’s insanely fast global adoption, and may be a route that Zuckerberg is similarly exploring.  The metaverse will require the purchase of significant hardware.  It, too, is an ambitious project that will launch decades after the Facebook flagship. Let’s all pay attention over the next couple of years and take notice when third-party developers – under a watchful eye and strict guidelines, of course – are invited to curate and broaden the metaverse experience in various ways, like shopping, gaming, utilities, fitness, and others.

Other tech CEOs have also profited marvelously in various ways on and off the Internet, and have pivoted to hardware in the process. Brin and Page monetized consumer intent with paid search advertising. Then they sold us Pixel phones and Google Home and acquired Nest for broader reach with devices. (And they’re betting big on Waymo.) Musk made his money online with PayPal when it sold to EBay, then monetized major hardware with Tesla electric cars. Bezos is a retailer and monetizes markup. He also likes hardware – Kindle and Echo both do just fine, thank you very much. With Meta, Zuckerberg seeks to do all of the above, just in the opposite order. He’ll first sell hardware to access the metaverse. Then he’ll sell advertising (likely highly contextualized) with a new model that combines search history, affinity, and basic demographics to a mostly Gen Z audience. He’ll build in some exchange system (maybe crypto-based) in the metaverse that costs real offline dollars. And he’ll most definitely build some kind of online shopping component.

So…what color would you prefer for your new goggles?

CONTENT CATEGORIES: fuel your funnel with the right stuff

Funnel marketing is back in fashion, and dozens of new ideas are popping up around this classic business concept.  The marketing funnel (or the consumer journey, or the conversion pathway, as it goes by many names,) is simply a conceptual construct to illustrate the broad phases of how leads are generated and then move from one place – where a consumer is typically unaware of your concept or brand or product – to another, more desirable place, where that consumer is ready to buy (or recommend) your specific product, and hopefully, right now.

It usually looks something like this:

Another concept – content marketing – has also been blogged to death in the last several years. The basic idea here is that brands can and should be generating and distributing a constant stream of content in many forms to attract and retain their target audiences.

Both funnel marketing and content marketing are relevant and valuable. Both concepts are basically applicable for just about any audience in just about any category, and that includes b-to-c or b-to-b. (That’s neat.)

But what I’ve noticed recently is that there isn’t much discussion on how content and marketing funnels can or should work together.

If you decide that it’s time to create content for your brand (and yes, it’s always a good time to do that,) you may also realize it can be quite difficult.  Questions abound about what to create, when to create it, where and how to distribute it, and whether or not it’s a good investment of critical resources, such as time, talent, and capital.

One important strategy is to create categories for your content that line up with your funnel marketing goals. In this case, the illustration would look more like this.

Let’s explore.

Almost every marketing funnel is illustrated by the letters A-I-D-A to represent awareness, interest, desire (or decision) and action. But in broader terms, the consumer journey is encapsulated by three broad categories: evaluation, consideration, motivation.

When the consumer is in the evaluation stage (becomes aware, develops interest):

In the early (high-funnel) stages, a consumer may be evaluating a purchase or interaction in this category.  In some cases, that consumer may be wholly unaware of your brand at this point. If your brand is new, or has just launched a new feature, or a new line, or has been dormant for a while and is back in some way, you want to communicate to the consumer set that you have options that might be worthy of review. 

For this I recommend creating generalized content.  Think about ways to show the consumer simply that your brand belongs in the category and has something interesting (or better yet different) to offer. This is a great time to educate/inform the consumer.

Some examples would include social content, blog posts, listicles, product reviews (to outline the basic brand/product traits.)

When the consumer is in the consideration stage (has interest, develops desire):

In the middle (mid-funnel) stages, that same consumer has probably become a bit more educated as a result of their exploration, and they’re now considering which options are the best for him/her/them. Note that this can only occur with brands that the consumer is aware of, and knows something about. They may circle back and look to check off important boxes, such as features, availability, time to delivery, and other (buying signal) particulars that are now important to them. 

Also note how the consumer, from a psychological perspective, gets more and more self-interested as they proceed down the funnel.  The conversation tends to move from “what does this thingy do?” to “what does this thingy do FOR ME?” 

This is where you should consider more specialized content. Help the consumer see your brand from the perspective of its superiority points, or better yet its unique points.

Some examples would include infographics to position your brand in the category, video or animated product demonstrations, info sheets/brochures or White Papers for b-to-b (to highlight the brand/product difference relative to other choices.)

When the consumer is in the action stage (has desire, ready to act):

Finally, the consumer reaches the moment of truth.  They’ve moved into the mode of desire and are ready to act in some way.  As mentioned above, they’ve considered this from a fairly self-interested point of view, and come to believe that maybe only one brand can really satisfy their needs.  Very often, they may whittle their choice down to two brands (because binary choices are easier for consumers to make,) and run a final A vs B competition in their minds, and yes, maybe even in their hearts.

Whenever you hear someone say marketing is emotional, they’re talking about this critical juncture in the funnel.  Consumers – especially if they’ve reached this binary choice phase – tend to go with the option that “feels” right for them.

If you’ve made it this far, be sure to have some contextualized content ready to go to motivate that consumer to choose your brand.  Take out any remaining guesswork. Show that consumer what it will be like to interact with your brand every day, and how that relationship will progress.

Talk about expectations, get specific on policies and procedures, warranties and registrations, and smooth the path to get the action (which may not always be a sale, by the way) you desire most.

Some examples would include consumer testimonials, case studies, and any customized or educational content like webinars (note the very personalized and specific complexion of these options.)

A clash of cultures: Twitter cancels Burger King.

A lot has been made of Burger King’s recent ad titled “Women Belong in the Kitchen.” If you’ve heard about it, you’ve likely already taken sides and are either itching to rage-tweet me, or are eager to hear someone else who supports your point of view.  Instead of taking sides, let’s be objective and unpack this thing one step at a time.

For anyone who doesn’t know, or didn’t read past the headline, Burger King was announcing the establishment of a new scholarship called H.E.R. (Helping Equalize Restaurants) to aid aspiring female employees who want to pursue careers as Chefs. The timing of its release coincided with International Women’s Day. 

Here’s the ad that ran as a full page in The New York Times:

First, let’s clarify what the ad was meant to do.  And we can do so by remembering what ALL ads are meant to do: get your attention. And this headline, while controversial if it stood alone, does that very well, because it’s dangerous. Because it’s a trope. Advertising leverages drama because it leads the reader to a destination that’s equal parts entertaining and attention-getting. And because a headline that reads “Burger King launches new scholarship to aid female representation in restaurant kitchens” is neither.  That sort of thing is for a press release, not an advert.

From a craft point of view, this is a strong headline, in that it serves to do at least one job that all good headlines should perform: it summarizes the content that follows. If we’re being objective, (and we agreed that we would be,) this is a very good all-copy ad-nouncement.

Now, let’s look at where it went wrong: in a word, Twitter. When the brand (and the agency behind it,) wanted to extend this exciting conversation online, it took to Twitter and Burger King’s 1.9 million followers with the initial tweet. Which, sadly, was just the headline. It then tweeted a summary of the content that follows in the ad. [Important note: the tweets were initially “debated” on @BurgerKingUK.] While Burger King did clarify the headline tweet in subsequent posts, it was apparently the string of ugly comments in the conversation thread that got out of control. The entire thread has since been deleted, and an apology was issued by global CMO Fernando Machado.

Ad culture meets Twitter culture and fails.  Cancel culture meets Burger King and shuts it down. This whole thing has gotten off the rails, and I think it’s mostly because people are not taking anything beyond face value. I would argue that we need a context culture more than anything else these days.  An army of fact checkers and industry experts who could act as docents for a whole generation of people who seem to crave being offended, and who magically find a fix on social media at roughly the rate of every news cycle.

The ad, the subsequent Twitterstorm, and the media kerfuffle that followed it have become new facets in the cultural touchstone that is today’s cancel-happy culture. The sad part is, it’s a pretty good ad. And Burger King, as a restaurant chain, (whether we should call them a “restaurant” or not is a different subject altogether,) is trying in earnest to do a darn good thing in the face of an inequality on which they are wholly qualified to comment. It’s a shame that we’re dealing with this level of bullshit from a minority of wokesters when a brand decides to put its money into something that might actually help in a concrete way what is, in this case, a marginalized segment of the population.

Now let’s look at what’s REALLY wrong with this ad: the typesetting is insulting, and should be cancelled immediately! The face is what it is – Burger King’s going for the retro-hip thing with the old bubble letters logo. Fine. I’ll concede that for the sake of the old-is-new branding mission.

But lord, where is the copyfitting? When the creative director was reviewing this, didn’t he or she think, “hmmm…that’s a weird place for a hyphen?” In the middle of the name of your new scholarship, in the middle of what’s arguably the most important word (Equalize,) you couldn’t break the line differently? And then again, in the last line of the ad, in another important word (kitchen) we couldn’t hard kern a little bit?

After a week of debating the merits of this approach, I haven’t heard any ad geeks talking about this.  Why? If we’re being objective, there’s probably a conspiracy afoot.